EP4587228A1

EP4587228A1 - Humanoid robot with tracks on legs configured for stand up sequence

Info

Publication number: EP4587228A1
Application number: EP23818365.1A
Authority: EP
Inventors: Lukas HEINZLE; Pascal Gohl; Roman Steffen; Mina Samir Fekry Kamel
Original assignee: Hexagon Technology Center GmbH
Current assignee: Hexagon Robotics GmbH
Priority date: 2023-12-04
Filing date: 2023-12-04
Publication date: 2025-07-23
Also published as: WO2025119448A1

Abstract

The invention relates generally to a mobile robot configured to provide metrology grade measurements in production environments, e.g. to generally support quality control, and especially lengthy, high-precision measurement or calibration activities that require long-term, concentrated effort and/or non-natural bodily gestures by a human operator. Aspects of the mobile robot, inter alia, relate to details regarding a hybrid track drive, a hip joint mechanism, tactile sensing, and workflows for teaching the robot different behaviors. In particular, benefits of mobility provided by mobile robots are transformed to the field of metrology while maintaining metrology grade accuracy.

Description

HUMANOID ROBOT WITH TRACKS ON LEGS CONFIGURED FOR STAND UP SEQUENCE

FIELD OF THE INVENTION

[0001] The invention relates generally to a mobile robot configured to provide metrology grade measurements in production environments. E.g., to generally support quality control, and especially lengthy, high-precision measurement or calibration activities that require long-term, concentrated effort and/or non-natural bodily gestures by a human operator.

BACKGROUND OF THE INVENTION

[0002] Conventional precise metrology devices need to be very stable in order to withstand inertial distortions that may arise due to its own operating weight and - especially since fast measurements are also desirable - its movements. Conventional devices are thus very heavy devices that are complicated to move and cannot be installed everywhere, e.g. due to weightloading restrictions. It would be beneficial to have the mobility and increased flexibility provided by mobile robots also in the field of metrology, which, for example, would reduce space required for the bulky and heavy metrology devices, increase flexibility in measuring different types of objects, and provide more efficient industrial processes. However, the structure of a mobile robot, e.g. including a lot of moveable parts and joints, introduces additional degrees of freedom, which make providing precise measuring more difficult. In addition, the mobility can lead to additional mechanical load on the sensitive metrology sensors, which must be taken into account.

[0003] Many conventional manufacturing environments are not designed to be barrier-free, which represents a challenge for wheeled or tracked movement by conventional robots. A walking movement on the other hand is typically slower, less energy efficient and/or causes increased internal vibration and accelerations. It would thus be desirable to provide a mobile robot, which can efficiently adapt to different environments, different operating modes, movement types and statures.

OBJECT OF THE INVENTION

[0004] It is therefore an object of the present invention to provide a mobile robot, which overcomes deficiencies of prior art robots in the field of quality control and metrology.

[0005] A further object is to provide increased mobility for the mobile robot as a whole and/or for the measurement implements of the robot. [0006] These objectives are at least partly achieved by realizing the characterizing features of the independent claims. Features, which further develop the invention in an alternative or advantageous manner are described in the dependent patent claims.

SUMMARY OF THE INVENTION

[0007] One aspect of the invention relates to a humanoid robot. The robot comprises a main body and two mechanically actuated legs attached to the main body at a lower portion of the main body by respective upper leg joints located at opposing sides of the main body. By way of example many aspects of the invention are discussed in details, respectively illustrated, via humanoid robots. Nevertheless, the applicability of the aspects of the inventions, unless explicitly provided, are not limited to humanoid robot. Similarly, terms such as "leg," "arm," "head," etc., are not to be construed as representing structurally identical, or not even similar, components of a human body part.

[0008] Each of the legs of the robot can be swiveled independently of the other leg about a respective swivel axis of the corresponding upper leg joint. Each of the legs comprises corresponding upper and lower parts, which are connected to each other via a lower leg joint and can be swiveled against each other about an axis of the lower leg joint. The upper leg joint might also be referred to as "hip joint", while the lower leg joint might also be referred to as "knee joint".

[0009] Each of the lower parts comprises a track drive comprising a track running over a lower pulley and an upper pulley. The track drive provides a running surface between the lower and the upper pulley. The lower pulley is arranged at a distal end away from the lower leg joint and the upper pulley is arranged closer to the lower leg joint than the lower pulley.

[0010] Each of the lower leg joints is configured to provide 360° rotatability for the lower part about the axis of the lower leg joint and the robot is configured to provide a stand-up sequence. The stand-up sequence comprises (i) establishing an inverted state of the two lower parts, and (ii) raising the upper pulleys from the ground by swiveling the lower leg joints in opposing directions.

[0011] In the inverted state a first vector extending from the upper pulley to the lower pulley of one of the two lower parts and a second vector extending from the upper pulley to the lower pulley of the other of the two lower parts point to opposite sides of the robot. In some specific embodiments the first vector points to an area in front of the robot and the second vector points to an area in the back of the robot.

[0012] The swiveling of the lower leg joint might be realized in a passive manner, when the swiveling is a consequence of an action of another component of the track drive, e.g., by driving the track. The swiveling can also be actively realized by a drive element assigned to the joint causing a targeted swiveling movement, in particular independently of the other elements of the leg. The swiveling can also be realized as a combination of the above. The robot might select the type of stand-up action based on the internal or external conditions, in particular the ground conditions.

[0013] In some embodiments in the inverted state, the first vector comprises a forward pointing component and the second vector comprises a backward pointing component, defined with respect to (i) a central body axis disposed between the upper leg joints and extending in a direction perpendicular to a leg suspension axis connecting the upper leg joints, and (ii) a middle plane comprising the leg suspension axis and the central body axis. The forward and backward pointing components are perpendicular to the middle plane and point to opposite directions.

[0014] In some specific embodiments in the inverted state each of the lower pulleys is farther away from the central body axis than the corresponding upper pulley.

[0015] In some specific embodiment in the inverted state one of the lower parts is completely located in a front domain and the other lower part is completely located in a rear domain. The front domain is the domain on one side of the middle plane and the rear domain is the domain on the other side of the middle plane.

[0016] In some specific embodiments in the inverted state each of the lower parts is partly located in a front domain and partly located in a rear domain, wherein the front domain is the domain on one side of the middle plane and the rear domain is the domain on the other side of the middle plane. One upper pulley and one lower pulley might be located in the front domain and one upper pulley and one lower pulley might be located in the rear domain.

[0017] In some embodiments the axes ofthe lower leg joints are parallel to each other. In some specific embodiments the axes ofthe lower leg joints are aligned to leg suspension axis connecting the upper leg joints.

[0018] In some embodiments the robot is configured to provide locomotion by track locomotion both in a stand-up mode and in a full track mode. The stand-up mode provides a surface contact face of the running surface that has smaller area size than a surface contact face of the running surface provided by the full track mode. In the stand-up mode the robot automatically arranges the lower part relative to the upper part such that during the locomotion the upper pulley is raised to a raised position that is farther from ground than a position of the lower pulley.

[0019] In some specific embodiments during the stand-up sequence the robot supports itself by self-balancing locomotion to maintain the raised position of the upper pulley. In particular the stand-up sequence ends in a stand-up position where the robot supports itself by solely standing on a curved part of the track which is curved by the circumferential area of the lower pulley.

[0020] In some specific embodiments the robot comprises an inertial sensor, a gyroscopic sensor and a control algorithm. The control algorithm is configured to automatically control the track drive based on the inertial sensor such that the robot is balanced in a defined upright position associated with the raised position of the upper pulley. The inertial sensor might be embodied as a set of accelerometers and the gyroscopes.

[0021] A further aspect of aspect of the invention, taken separately or in combination with the other aspects of the invention, relates to a humanoid robot. The robot comprises a main body and mechanically actuated extremities attached to the main body. The mechanically actuated extremities comprise two mechanically actuated legs attached to the main body at a lower portion of the main body by respective upper leg joints. Each of the legs can be swiveled independently of the other leg about a swivel axis of the corresponding upper leg joint. The legs are configured to provide locomotion of the robot over ground. At least one of the mechanically actuated extremities is releasably attachable to the main body by an extremity locking mechanism.

[0022] The extremity locking mechanism is provided by a receptacle having a depression for receiving a spigot along a penetration axis. The receptacle and the spigot have matching cornered, particularly hexagonal, shapes. The spigot has at least three, particularly six, latching elements. In a basic position of a release mechanism, each of the latching elements pushes radially outwards in order to engage in a corresponding cavity of the receptacle. An activation of the release mechanism enables the latching elements to radially escape into the spigot in order to allow the spigot to be released from the receptacle.

[0023] In some embodiments each of the latching elements is configured as a rotation body, in particular as a sphere or an ellipsoid, a trapezoid, a pyramid, a trapezoid having rounded corners, or a pyramid having rounded corners.

[0024] In some embodiments the latching elements and the cavities are configured and matched to each other in such a manner that the engagement of the latching elements in the cavities causes a self-centering of the spigot, in particular a self-centering with respect to the penetration axis.

[0025] In some embodiments the release mechanism is arranged in the spigot. The release mechanism comprises (i) at least one radial pin for activating the release mechanism, (ii) an axial pin for blocking or allowing radial escape of the latching elements, and (iii) a tensioning spring to maintain the basic position. The radial pin, the axial pin and the tensioning spring are operatively connected in such a way that (a) in the basic position of the release mechanism the axial pin forces the latching elements radially outwards, and (b) when the release mechanism is activated a displacement of the radial pin moves the axial pin towards the tensioning spring, and the axial pin releases space due to its displacement and thus enables the radial escape of the latching elements into the spigot.

[0026] In some specific embodiments in the basic position of the axial pin presses the latching elements into the cavities by means of a tensioning force due to the tensioning spring. [0027] In some embodiments each of the latching elements has at least two points of contact with its corresponding cavity.

[0028] In some embodiments the extremity locking mechanism is configured to provide (i) a mounted state, and (ii) a dismounting functionality. In the mounted state at least one of the mechanically actuated extremities is mechanically constrained by the degrees of freedom provided by the joint attaching the at least one of the mechanically actuated extremities to the main body. In the mounted state a flow of electrical energy is enabled between the at least one of the mechanically actuated extremities and the main body. The dismounting functionality is configured to block the flow of electrical energy between the at least one of the mechanically actuated extremities and the main body. The dismounting functionality is also configured to cancel at least a part of the mechanical constraints between the at least one of the mechanically actuated extremities and the main body. I.e. the dismounting functionality allows a separation of the at least one of the mechanically actuated extremities from the main body.

[0029] In some specific embodiments the extremity locking mechanism comprises a set of pogo-pins, or in other words spring loaded electric connectors comprising an integrated helical spring in the pin, arranged on a front face of the spigot and a set of corresponding recesses arranged on a corresponding face of the depression in the receptable. The pogo-pins and the recesses are configured to engage each other thereby providing an electrical connection between the mechanically actuated extremity and the main body. The integrated helical spring in the pin applies a constant normal force against the recess. Such pogo-pins are therefore especially suited for providing stable electrical connection between moving components.

[0030] In some embodiments the receptacle and the spigot have shapes that only allow inserting the spigot into the receptacle in a single pre-defined relative orientation with respect to each other. In some specific embodiments the engagement of the latching elements in the cavities causes a self-alignment of corresponding rotational angles in a plane perpendicular to the penetration axis, in particular the self-centering and self-alignment provides a fixed initial angle for the leg.

[0031] A further aspect of aspect of the invention, taken separately or in combination with the other aspects of the invention, relates to a humanoid robot. The robot comprises a main body and two mechanically actuated legs attached to the main body at a lower portion of the main body by respective upper leg joints located at opposing sides of the main body. Each of the legs can be swiveled independently of the other leg about a respective swivel axis provided by the corresponding upper leg joint.

[0032] Each of the legs comprises corresponding upper and lower parts, which are connected to each other via a lower leg joint. The lower and upper parts and can be swiveled against each other about an axis provided by the lower leg joint. Each of the lower parts comprises a track drive comprises a track running over a lower pulley and an upper pulley. The track drive provides a running surface between the lower and the upper pulley. The lower pulley is arranged at a distal end away from the lower leg joint and the upper pulley is arranged closer to the lower leg joint than the lower pulley.

[0033] The robot is configured to angle the lower parts relative to the upper parts to support the robot on the running surfaces to provide locomotion by the track drive. The robot is configured to provide a stand-up position. In the stand-up position the lower parts are raised relative to the upper parts such that the upper pulleys are raised to raised positions that are farther from ground than positions of the lower pulleys. In the stand-up position the running surfaces might be vertical.

[0034] Each of the track drives comprises a pretensioning arrangement. The pretensioning arrangement is configured to provide tensioning of the track by providing a tensioning force acting to increase a distance between the lower pulley and the upper pulley. The pretensioning arrangement is configured to set the tensioning force by making use of a gravitational force and/or an impact force acting on the pretensioning arrangement in the stand-up position.

[0035] In some embodiments each of the pretensioning arrangements comprises a counter mechanism configured to set a tensioning state of a spring-based tensioning unit in reaction to the gravitational force and/or the impact force. The tensioning state of the spring-based tensioning unit might be provided in incremental steps as a function of a tension of the track.

[0036] In some specific embodiments the spring-based tensioning unit comprises an upper fixed component, a lower fixed component, and a floating component. The floating component is arranged between the upper and lower pulley, the upper fixed component is fixed with the upper pulley, and the lower fixed component is fixed with the lower pulley. The counter mechanism is provided by a latch arrangement connecting the floating component with the upper and lower fixed components. A movement of the floating component relative to the upper fixed component and movement of the floating component relative to the lower fixed component are restricted to different unidirectional directions of movement. A relative position of the floating component and one of the upper or lower fixed components defines the tensioning state of the spring-based tensioning unit.

[0037] In some specific embodiments the spring-based tensioning unit comprises two springs. A first of the two springs connects the floating body and the upper fixed component and a second of the two springs connects the floating body unit and the lower fixed component. The latch arrangement comprises a first latch component between the floating body and the upper fixed component. The first latch component is configured to restrict a movement of the floating body with respect to the upper fixed component to a first unidirectional movement. The latch arrangement comprises a second latch component between the floating body and the lower fixed component. The second latch component is configured to restrict a movement of the floating body with respect to the lower fixed component to a second unidirectional movement. The first and second unidirectional movement have opposite directions. [0038] In some specific embodiments the pretensioning arrangement comprises a rope connecting the second of the two springs and the lower fixed component. The rope transmits a spring force of the second of the two springs to the lower fixed component. The upper fixed component has a rope guide surface, in particular in the form of a surface with a semicircular crosssection, which acts as a fixed deflection pulley with respect to the rope and thereby converts the direction of pull of the rope.

[0039] In some specific embodiments the second spring is high stiffness spring. High stiffness in the sense of the invention means that the spring experience no significant deformation owing to the weight of the robot and/or the weight of the track drive. In other words, the spatial relations of the members of the track drive is invariant irrespectively whether track drive is horizontal, vertically hanging on the upper pulley or vertically supporting the weight of the robot.

[0040] In some embodiments the pretensioning arrangement comprises and axial member and a peripherical member. The axial member arranged along an axis connecting the upper and lower pulleys, such that is axially displaceable along said axis. The peripherical member is arranged transversally offset from said axis, such that it is transversally displaceable by the axial displacement of the axial member. The peripherical member is configured to set by its transversal displacement a distance between the upper and lower fixed components. In some specific embodiments the axial member has a conical or frustrum shape, and the peripherical member has a conical or frustrum shape. In some specific embodiments the floating body comprises a further peripherical member arranged symmetrically to the peripherical member.

[0041] In some embodiments the tensioning of the tracks causes portions of the track located between the upper and lower pulleys to align to a common tangent of said pulleys.

[0042] A further aspect of aspect of the invention, taken separately or in combination with the other aspects of the invention, relates to a humanoid robot. The humanoid robot comprises a main body and two mechanically actuated legs attached to the main body at a lower portion of the main body by respective upper leg joint arrangements located at opposing sides of the main body.

[0043] Each of the legs can be swiveled independently of the other leg about a respective first swivel axis of the corresponding upper leg joint arrangement for deflecting the legs forwards and backwards for providing a walking motion of the robot. The first swivel axes and an axis connecting the upper leg joint arrangements might be located in the same plane. Each of the legs comprises a track drive comprising a track providing a running surface for track locomotion.

[0044] Each of the upper leg joint arrangements is configured to provide a pivotability for the corresponding leg about a second swivel axis of the upper leg joint arrangement that is different from the respective first swivel axis. The second swivel axis might be orthogonal to the axis connecting the upper leg joint arrangements. A respective motor unit for providing the pivotability for the corresponding leg about the second swivel axis is arranged offset from the leg joint arrangement. [0045] The motor unit comprises a linear actuator. The linear actuator might be arranged in the main body. The motor unit comprises (i) an axle, (ii) a traveler configured to be linearly displaceable along the axle, and (iii) a motor to move the traveler along the axle. The traveler is kinematically linked by a transfer element to a rotary element of the upper leg joint arrangement for providing the pivotability of the leg about the second swivel axis. The transfer element and the rotary element are configured to translate a linear movement of the traveler to a rotational movement of the rotary element.

[0046] In some embodiments the humanoid robot is configured to provide a walking locomotion by shifting its center of gravity by a coordinated swiveling of the legs about the corresponding second swivel axis.

[0047] In some embodiments each of the linear actuators comprises at least one passive damping element. The passive damping element is arranged to an endpoint of the axle and configured to provide a restore force acting on the traveler and a causing a departure of the traveler from the endpoint. The passive damping element might be embodied as a spring.

[0048] In some embodiments the linear actuator comprises a ball-screw gear.

[0049] In some embodiments a displacement range of each of the second swivel axes is at least 10°. The second swivel axes might be configured to provide an inward tilt of the corresponding legs.

[0050] In some embodiments each of the leg joint arrangements comprises a further rotary element for providing a swiveling of the respective leg about the respective first swivel axis, Said further rotary elements are located outside a volume defined by a housing of the main body.

[0051] In some embodiments the robot is configured to maintain a defined orientation of the main body with respect to a direction of gravity by the coordinated swiveling of the legs about the corresponding second swivel axis.

[0052] In some embodiments the robot is configured maintain contact between the ground and each of the running surfaces by a coordinated swiveling of the legs about the corresponding second swivel axis.

[0053] A further aspect of aspect of the invention, taken separately or in combination with the other aspects of the invention, relates to a mobile robot. The robot comprises a main body and two mechanically actuated legs attached to the main body at a lower portion of the main body by respective upper leg joints.

[0054] Each of the legs comprises respective first and second track drives. Each of the track drives comprising a track running over respective lower and upper pulleys and providing a corresponding running surface for track locomotion. The upper pulleys of the first and second track drives of the first leg are mounted coaxially to a track swivel axis of the first leg, such that the first and second track drives of the first leg are independently pivotable about the track swivel axis of the first leg. The upper pulleys of the first and second track drives of the second leg are mounted coaxially to a track swivel axis of the second leg, such that the first and second track drives of the second leg are independently pivotable about the track swivel axis of the second leg.

[0055] In some embodiments each of the legs comprises corresponding upper and lower parts, which are connected to each other via a lower leg joint such that they can be swiveled against each other. The respective first and second track drives are comprised by the lower parts of the corresponding leg. The respective track swivel axes are provided by the corresponding lower leg joints.

[0056] In some specific embodiments the first and second track drives are located on an outer face of the upper part of the leg, wherein the outer face is the face more remote from the main body.

[0057] In some embodiments the robot is configured to provide an extended track mode. In the extended track mode for each of the legs the respective first and second track drives are twisted to each other about the respective track swivel axis so that the upper pulleys are located between the lower pulleys. In the extended track mode track locomotion is provided by running surfaces of all the first and second track drives. The upper and lower pulleys of the first and second track drives might be arranged to a common plane.

[0058] In some specific embodiments the robot is to provide an extended track stand-up sequence. The extended track stand-up sequence comprises (i) establishing the extended track mode for both of the legs, and (ii) raising the upper pulleys from the ground by swiveling each of first and second track drives in a coordinated manner such that the robot supports itself on the four lower pulleys. The specific features of the stand-up sequence might be applied correspondingly to the extended stand-up sequence.

[0059] In some embodiments the robot is configured to provide a climbing mode. In the climbing mode an angle between a first track vector extending from the upper pulley of the first track drive to the lower pulley of the first track drive and a second track vector extending from the upper pulley of the second track drive to the lower pulley second track drive falls into a range of 15° to 55°.The skilled person understands that the features of the extended track mode and the climbing mode are beneficially combinable with each other.

[0060] In some specific embodiments the robot is configured to adjust the angle between the first and second track vectors such that both the first and second track drives maintain at least two- point contact with the ground.

[0061] In some embodiments for each of the legs, the respective leg and at least one of the respective first track drive and the respective second track drive comprise matching counterparts of a coupling interface configured to releasably attach the at least one of the respective first track drive and the respective second track drive to the respective leg. The coupling interface is configured to provide for independent swiveling of the first track drive and the second track drive about the track swivel axis.

[0062] In some specific embodiments the coupling interface is configured to releasably attach the respective first track drive and the respective second track drive to opposite sides of the respective leg.

[0063] In some specific embodiments the coupling interface is configured to releasably attach the respective second track drive to the respective first track drive. The respective leg and the respective first and second track drives are stacked one above the other in the order leg-first track drive-second track drive so that they can rotate around the respective track swivel axis of the leg. The first and second track drives are independently pivotable about the track swivel axis of the leg.

[0064] In some specific embodiments an additional interface part is passed through the first track drive into the leg axially to the track swivel axis. At a distal end inside the leg the additional interface part is configured to engage a rotating motor component inside the leg so that a rotational movement about the track swivel axis is transmitted from the rotating motor component inside the leg to the second track drive by means of the additional interface part. At a distal end on side of the second track drive the second track drive is attached with its counterpart of the coupling interface to a matching counterpart on the additional interface component. Thus, the engaging with the counterpart on the additional interface component provides for a swiveling of the second track drive about the track swivel axis.

[0065] In some specific embodiments the first track drive and the second track drive comprise identical counterparts of the coupling interface, each matching the counterpart of the coupling interface arranged on the respective leg, which is identical to the counterpart arranged on the additional interface component.

[0066] In some embodiments each of the first and second track drives comprises a corresponding track drive motor configured to provide a driving force for at least one of the respective lower or upper pulleys. Each of the legs comprises a first and a second swivel motor. The coupling interface is configured such that the first swivel motor drives a swiveling of the first track drive about the track swivel axis and the second swivel motor drives a swiveling of the second track drive about the track swivel axis.

[0067] In some specific embodiments the coupling interface is configured such that the second swivel motor drives two shafts. The two shafts are arranged and configured to engage coupling interface components that are approached to the leg from opposite sides of the leg.

[0068] In some embodiments the coupling interface is provided by a receptacle having a depression for receiving a spigot along the track swivel axis. The receptacle and the spigot have matching cornered, particularly hexagonal, shapes. The spigot has at least three, particularly six, latching bodies, wherein in a basic position of a release mechanism, each of the latching bodies pushes radially outwards, away from the track swivel axis, in order to engage in a corresponding cavity of the receptacle. An activation of the release mechanism enables the latching bodies to radially escape into the spigot in order to allow the spigot to be released from the receptacle.

[0069] The skilled person understands that the specific features of extremity locking mechanism are correspondingly applicable to the coupling interface.

[0070] A further aspect of aspect of the invention, taken separately or in combination with the other aspects of the invention, relates to a humanoid robot. The robot comprises a main body, two mechanically actuated legs attached to the main body at a lower portion of the main body by respective upper leg joints, and a mechanically actuated arm attached to the main body by an upper arm joint. The legs and the arm can be swiveled independently of one another. The humanoid robot comprises a depth measuring sensor, configured to generate distance measuring data to an environment of the robot.

[0071] The humanoid robot is configured to use the arm to hold and position a mobile scanning device in a plurality of positions and orientation relative to an object to be measured. The mobile scanning device is configured to approach the object to be measured and to generate 3D position measuring data of the object to be measured. In particular, the robot is configured to hold the device with a position and orientation accuracy which corresponds to a calibration routine of said device. More particularly the robot is configured to hold the mobile scanning device in an essentially vibration free manner.

[0072] The humanoid robot is configured to autonomously move the mobile scanning device (by moving the arm and/or the robot position) to different positions and orientations relative to the object to be measured, and based thereof, to use the mobile scanning device to generate 3D position measuring data of the object to be measured according to a defined measurement criterion. The humanoid robot is configured to determine and set a next position and orientation of the mobile scanning device by moving the arm based on an evaluation regarding achievement of the measurement criterion taking into account previous positions and orientations of the mobile scanning device, known measurement characteristics of the mobile scanning device, and distance measuring data of the depth measuring sensor for measuring distances to the object to be measured and the mobile scanning device.

[0073] In some embodiments the measurement criterion is at least one of (i) a point density, (ii) a distance measurement accuracy, (iii) a coordinate measurement accuracy, and (iv) a signal- to-noise ratio.

[0074] In some embodiments the mobile scanning device is embodied as a laser-based scanner, a tactile scanning device, or a stereo imaging device.

[0075] In some embodiments the depth measuring sensor is embodied as a laser-based distance measuring device, a time-of-flight camera, or a stereo imaging device. 72

[0076] In some embodiments the evaluation is configured to provide an optimization of a scanning distance and/or a viewing angle of the mobile scanning device to the object to be measured in order to meet the measurement criterion.

[0077] In some embodiments the robot is configured to determine the previous positions and orientations of the mobile scanning device by tracking position and orientation of the mobile scanning device by the distance measuring data of the depth measuring sensor.

[0078] In some embodiments the robot is configured to determine the previous positions and orientations of the mobile scanning device by tracking position and orientation of the mobile scanning device based on position data of a kinematic chain of moving parts of the robot, in particular the arm and the legs of the robot.

[0079] In some embodiments the mobile scanning device is configured to be tracked by a tracking device, particularly an industrial laser tracker. The humanoid robot is configured to determine the next position and orientation of the mobile scanning device based on distance measuring data of the depth measuring sensor for measuring distances to the mobile scanning device and the tracking device, by evaluating relative positions and orientations of the mobile scanning device relative to the object to be measured and the tracking device.

[0080] In some specific embodiments the humanoid robot is configured to determine track data for tracking position and orientation of the mobile scanning device relative to a position of the tracking device and the robot by using the distance measuring data of the depth measuring sensor and to use the track data to identify and avoid upcoming line-of-sight breaking between the tracking device and the mobile scanner.

[0081] In some embodiments the evaluation regarding achievement of the measurement criterion takes into account feedback from the mobile scanner.

[0082] A further aspect of aspect of the invention, taken separately or in combination with the other aspects of the invention, relates to a humanoid robot. The robot comprises a main body, a mechanically actuated arm attached to the main body, and a hand-joint arrangement at a distal end of the arm remote from the main body.

[0083] The hand-joint arrangement is configured to hold a touch sensor or the hand-joint arrangement comprises a touch sensor. The touch sensor comprises an elastomeric component for contacting a surface patch of an object to be measured. The touch sensor is configured to generate tactile sensing data that provide geometric 3D information of the surface patch based on observation of deformation of an inner face of the elastomeric component when an outer face of the elastomeric component is brought into physical contact with the object to be measured.

[0084] The robot is configured to carry out an iterative tactile sampling of different surface patches of the object with the elastomeric component, thereby generating tactile sensing data associated with the different surface patches. The iterative tactile sampling comprises movement of the arm and the hand-joint arrangement to change poses of the touch sensor to provide for touching the different surface patches with the elastomeric component, and generation of pose data that provide for referencing tactile sensing data associated with different poses of the touch sensor to a common coordinate system.

[0085] In some embodiments the pose data comprise kinematic chain data determined by angular encoders of the robot for tracking movement of the arm and/or the hand-joint arrangement.

[0086] In some embodiments the pose data comprise tracking data determined by a camerabased or light-based tracking sensor of the robot configured to capture at least part of at least one of the arm, the hand-joint arrangement, and the tactile sensor. In particular the camera-based or light-based tracking sensor is arranged in the main body or in a mechanically actuated head attached to the main body at the top of the main body. The camera-based or light-based tracking sensor might have an active illumination source included.

[0087] In some embodiments the robot is configured to provide the pose data by carrying out a point cloud matching procedure on tactile sensing data associated with overlapping surface patches of the different surface patches.

[0088] In some embodiments the robot is configured to use the tactile sensing data and the pose data to generate a 3D model of the object.

[0089] In some embodiments the robot comprises a visual sensor configured to generate visual data of the different surface patches. The robot is configured to provide for referencing the visual data relative to the tactile sensing data, particularly when the robot is configured to use the visual data to texture the 3D model of the object.

[0090] In some specific embodiments the robot is configured to use the visual data during the iterative tactile sampling to determine a next surface patch to be measured by the touch sensor.

[0091] In some specific embodiments the next surface patch is determined based on a surface reconstruction of the object by processing the visual data based on a photogrammetric principle, particularly based on a structure from motion algorithm.

[0092] In some specific embodiments the robot is configured to use the visual data to determine a tactile sensing confidence value of the tactile sensing data. The robot is configured to adapt placement of the touch sensor on the object during the iterative tactile sampling and/or to adapt a data assignment to complement the tactile sensing data with the visual data based on the tactile sensing confidence value.

[0093] In some specific embodiments the tactile sensing confidence value is determined by identifying flat surfaces and/or contrast changes exceeding a contrast change threshold.

[0094] In some embodiments the robot is configured to store and/or access an object database comprising digital 3D models. The robot is further configured to reference tactile sensing data with at least one digital 3D model from the object database. The robot might be configured to use said referencing during the iterative tactile sampling to determine a next surface patch to be measured by the touch sensor.

[0095] In some embodiments the robot is further configured (i) to derive a current pose of the front face of the elastomeric component, (ii) derive a shape and location of an object being touched based on the visual data, in particular using a structure from motion or simultaneous localization and mapping method, and (iii) to adopt an orientation of the arm and/or the hand-joint arrangement based on the shape and location of the object being touched and a current pose of the touch sensor.

[0096] In some embodiments the visual sensor is a low focal length camera arranged to the hand-joint arrangement and/or on the arm in the proximity of the hand-joint arrangement.

[0097] In some embodiments the touch sensor comprises a touch sensor interior volume and a touch sensor imaging arrangement disposed within the touch sensor interior volume. The touch sensor interior volume at least partly filled with the elastomeric component. The inner face the elastomeric component has a reflective coating. Touching of an object with the outer face causes at least a partial reproduction of a geometric shape of a portion of the touched object on the inner face. The touch sensor imaging arrangement is configured to provide a set of images from the objects touched by the touch sensor by imaging the inner face of the elastomeric component.

[0098] A further aspect of aspect of the invention, taken separately or in combination with the other aspects of the invention, relates to a humanoid robot configured for 3D surveying of an object. The robot comprises a main body, two mechanically actuated legs attached to the main body at a lower portion of the main body, a mechanically actuated arm attached to the main body, and a perception sensor. The legs and the arm can be swiveled independently of one another and the perception sensor is configured to generate perception data for perceiving a relative positional relationship between the object and at least part of the robot. The robot comprises a large language model module, in the following referred to as LLM-module. The LLM-module is configured to process text input of a planned surveying task to be executed by the robot, wherein the planned surveying task involves a specific movement of the robot arm relative to the object (e.g. wherein the robot arm guides and operates a metrology-grade sensor, which, for example, is part of the robot or is held by the robot) and the LLM-module is trained to transform the text input of the planned surveying task to an output comprising robot command language. By way of example, the LLM-module is provided by a (pre-)trained artificial neural network. The robot further comprises a robot arm instruction interpreter, a robot arm instruction executor, and a comparison module. The robot arm instruction interpreter is configured to select robot commands out of the output of the LLM-module. The robot arm instruction executor is configured to cause the robot to execute the robot commands selected by the robot arm instruction interpreter, resulting in an executed surveying task executed by the robot. The comparison module is configured to use perception data of the perception sensor capturing a relative positional relationship between the object and at least part of the robot, e.g. the arm, to compare the planned surveying task and the executed surveying task to provide a confidence value regarding achievement of the planned surveying task, wherein the robot is configured to take into account the confidence value to adapt movement of the robot when repeating the planned surveying task.

[0099] In one embodiment, the planned surveying task involves the robot to use a metrologygrade surveying sensor, wherein the metrology-grade surveying sensor is part of the robot or is held by a hand-joint arrangement of the robot arranged at a distal end of the arm remote from the main body.

[0100] In a further embodiment, the confidence value is provided based on measurement data of the metrology-grade surveying sensor.

[0101] In a further embodiment, the robot comprises an input functionality to provide user input to change the output of the LLM-module. Here, the robot further comprises (i) a display, wherein the robot is configured to provide - on the display - feedback to an operator of the robot based on the output of the LLM-module, and/or (ii) a text-to-speech module configured to acoustically provide feedback to an operator of the robot based on the output of the LLM-module.

[0102] In a further embodiment, the robot comprises a speech-to-text module configured to translate voice input to the text input of the planned surveying task, e.g. wherein the speech-to-text module is part of the LLM module.

[0103] In a further embodiment, the robot commands in robot command language comprise an instruction in a parse-able form, e.g. JavaScript Object Notation or extensible markup language.

[0104] In a further embodiment, the comparison module is configured to (i) analyze the text input of the planned surveying task to provide an input comparison parameter indicative of a movement of the robot involved by the planned surveying task, (ii) analyze the perception data to provide an output comparison parameter indicative of a movement of the robot involved by the executed surveying task, and (iii) compare the input comparison parameter and the output comparison parameter to provide the confidence value.

[0105] In a further embodiment, the robot is configured to take into account the confidence value to adapt processing of the LLM-module to provide a further output associated with the text input of the planned surveying task comprising robot commands in robot command language, wherein the robot is configured that the further output is processed by the robot arm instruction interpreter to provide a set of robot commands out of the further output. Alternatively, or in addition, the robot is configured to take into account the confidence value to adapt processing of the robot arm instruction interpreter to select robot commands out of the output or the further output of the LLM-module to provide a further set of robot commands out of the output or the further output. The robot is thus configured to carry out an adapted executed surveying task based on the robot arm instruction executor causing the robot to execute the set of robot commands and/or the further set of robot commands.

[0106] In a further embodiment, the comparison module is configured to analyze the text input of the planned surveying task to provide a planned position of the robot associated with the planned surveying task, e.g. a planned position relative to the object, and to compare the planned position and an actual position of the robot, e.g. an actual position relative to the object, to provide the confidence value. By way of example, a distance between the planned position and the actual position is determined to provide the confidence value.

[0107] In a further embodiment, the robot comprises a locomotion instruction interpreter, configured to select robot commands out of the output of the LLM-module that provide instructions for locomotion of the robot. The robot further comprises a robot locomotion instruction executor, configured to cause locomotion of the robot according to the robot commands selected by the locomotion instruction interpreter, wherein the locomotion of the robot is part of the executed surveying task.

[0108] By way of example, the robot is configured (i) to take into account the confidence value to adapt processing of the locomotion instruction interpreter to provide a selection of locomotion commands for a locomotion of the robot out of an output of the LLM-module, and (ii) to carry out an adapted executed surveying task based on the locomotion instruction executor causing the robot to execute the selection of locomotion commands.

[0109] In some specific embodiments the comparison module is configured to determine the actual position of the robot based on at least one of (i) data of an inertial measuring unit of the robot, (ii) image data provided by an imaging unit of the robot, and (iii) depth measuring data of a depth sensor of the robot. The perception sensor might be configured to provide at least one of the image data and the depth measuring data.

[0110] In some embodiments the robot is configured to stop operation of the robot as a function of a defined termination criterion for successively derived confidence values.

[0111] In some embodiments the comparison module is configured to provide the confidence value by determining an actual pose of at least part of the robot and comparing the actual pose with a list of known poses of the at least part of the robot and a corresponding defined range of a permissible pose deviation for each of the known poses.

[0112] In some embodiments the comparison module is configured to provide the confidence value by determining an actual movement pattern of at least part of the robot and comparing the actual movement pattern with defined permissible movement patterns of the robot.

[0113] In some embodiments the comparison module is configured to provide the confidence value by determining a change in an environment of the robot before and after execution of the 77 executed surveying task. In particular by determining a position of at least part of the robot relative to the object and/or a change of positions of further objects in the environment.

[0114] In some embodiments the comparison module is configured to provide the confidence value based on an explicit feedback regarding the executed surveying task from an operator. In particular wherein at least a part of explicit feedback is provided as voice input from the operator.

[0115] A further aspect of aspect of the invention, taken separately or in combination with the other aspects of the invention, relates to a humanoid robot. The robot comprises a main body, two mechanically actuated legs attached to the main body at a lower portion of the main body, a mechanically actuated arm attached to the main body, a hand-joint arrangement at a distal end of the arm remote from the main body, and a perception sensor. The legs, the arm, and the handjoint arrangement can be moved independently of one another. The perception sensor is configured to generate visual perception data of an environment of the robot.

[0116] The robot is configured to execute surveying tasks that involve specific movements of the robot and the use of a metrology-grade surveying sensor. The metrology-grade surveying sensor is part of the robot or is held by the hand-joint arrangement.

[0117] The robot comprises a large language module, a gesture recognition module, an instruction interpreter, and a robot instruction executor.

[0118] The large language module (LLM module, see also above) is configured to transform text input of a planned surveying task to an output comprising robot command language. The gesture recognition module is configured to analyze the perception data to recognize a gesture made by an operator that is associated with the planned surveying task. The instruction interpreter is configured to provide robot commands based on output of the LLM-module and the gesture recognition module, and the a robot instruction executor is configured to cause the robot to execute the robot commands, resulting in an executed surveying task executed by the robot.

[0119] A further aspect of aspect of the invention, taken separately or in combination with the other aspects of the invention, relates to a humanoid robot. The robot comprises a main body, two mechanically actuated legs attached to the main body at a lower portion of the main body, a mechanically actuated arm attached to the main body, a hand-joint arrangement at a distal end of the arm remote from the main body, and a perception sensor. The legs, the arm, and the handjoint arrangement can be moved independently of one another. The perception sensor is configured to generate visual perception data of an environment of the robot.

[0120] The robot is configured to execute surveying tasks that involve specific movements of the robot and the use of a metrology-grade surveying sensor. The metrology-grade surveying sensor is part of the robot or is held by the hand-joint arrangement.

[0121] The robot comprises a gesture recognition module, an instruction interpreter, a robot instruction executor, and a confirmation module. [0122] The gesture recognition module is configured to analyze the perception data to recognize a gesture made by an operator that is associated with a planned surveying task that involves a specific movement of the robot and the use of the metrology-grade surveying sensor. The gesture recognition module is configured to translate a recognized gesture into a planned model movement sequence of a set of robot parts that are involved in the planned task.

[0123] The instruction interpreter is configured to provide robot commands based on the planned model movement sequence.

[0124] The robot instruction executor is configured to cause the robot to execute the robot commands, resulting in an executed surveying task executed by the robot.

[0125] The confirmation module is configured (i) to analyze perception data of the perception sensor capturing at least part of the set of robot parts to recognize an observed movement of the set of robot parts, (ii) to translate the observed movement into an executed model movement sequence of the set of robot parts, and (iii) to provide a confidence value regarding a matching of the planned model movement sequence and the executed model movement sequence. The robot is configured to take into account the confidence value to adapt movement of the robot when repeating the planned surveying task.

[0126] In some embodiments the robot is configured to take into account the confidence value to adapt processing of the gesture recognition module to provide a further planned model movement sequence. The robot is configured that the further model movement sequence is processed by the instruction interpreter to provide a set of robot commands. Alternatively, or in addition, the robot is configured to take into account the confidence value to adapt processing of the instruction interpreter to provide a further set of robot commands based on the planned model movement sequence or the further model movement sequence. Therefore, the robot is configured to carry out an adapted executed surveying task based on the robot instruction executor causing the robot to execute the set of robot commands and/or the further set of robot commands.

[0127] In some embodiments the confidence value is provided based on measurement data of the metrology-grade surveying sensor and/or measurement data of the perception sensor relative to the environment.

[0128] In some embodiments the robot is configured to provide the planned model movement sequence as a scaled movement sequence or an absolute movement sequence. For the absolute movement sequence the magnitudes of the planned model movement sequence are in one-to-one correspondence with the recognized gesture made by the operator. For a scaled movement sequence at least one magnitude of the planned model movement sequence is rescaled with respect to recognized gesture made by the operator by the gesture recognition module.

[0129] In some specific embodiments the rescaling is carried out on the basis of the different proportions and/or joint mobility of the operator and robot. [0130] A further aspect of aspect of the invention, taken separately or in combination with the other aspects of the invention, relates to a humanoid robot configured for 3D surveying of an environment. The robot comprises a main body, two mechanically actuated legs attached to the main body at a lower portion of the main body, and a mechanically actuated arm attached to the main body at an arm joint at an upper portion of the main body. Each of the legs comprises a track drive providing a track for track locomotion, and the robot is configured to autonomously move through the environment and to autonomously change between an upright measurement mode and a crouching measurement mode.

[0131] In the upright measurement mode, the main body is placed higher above ground than in the crouching measurement mode, contact faces of the tracks contacting the ground have smaller area sizes than contact faces of the tracks in the crouching measurement mode, and a posture of the robot is automatically balanced based on a control algorithm configured to automatically stabilize the posture based on inertial data of an inertial sensor of the robot.

[0132] In the crouching measurement mode, the robot resumes a tilt-resistant posture by having increased area sizes of the contact faces of the tracks compared to the upright measurement mode, such that the tilt-resistant posture can be maintained free of active balancing by the robot.

[0133] The robot is configured to execute surveying tasks by a metrology-grade surveying sensor to generate 3D surveying data that provide geometric 3D information of the environment, wherein the metrology-grade surveying sensor is part of the arm or is held by the arm. The robot is configured to capture perception data of the environment and to carry out a classification of objects and environment areas within the perception data based on object type and area type, wherein different object types and different area types are assigned different surveying criteria for surveying with the surveying sensor. The robot is configured to use this classification to automatically switch between the upright measurement mode and the crouching measurement mode.

[0134] In one embodiment, the robot is configured to generate the 3D surveying data with motion correction based on compensation data of an inertial measurement unit, wherein a weighting of the compensation data for the motion correction is reduced in the crouching measurement mode compared to the upright measurement mode. In particular, the motion correction by the compensation data is deactivated in the crouching measurement mode.

[0135] In a further embodiment, the robot is configured that a range of motion of the surveying sensor relative to the robot is more restricted in the crouching measurement mode than in the upright measurement mode.

[0136] In a further embodiment, the robot is configured to provide at least one defined fixed holding position of the surveying sensor relative to the main body configured to provide increased mechanical stability of the surveying sensor compared to an arrangement of the surveying sensor out of the at least one fixed holding position. The surveying sensor is kept in the at least one fixed holding position when generating the 3D surveying data in the crouching measurement mode. For example, the at least one fixed holding position involves a (mechanical) locking of the surveying sensor at the main body or a (mechanical) locking of the arm at the main body at a locking point of the arm that differs from the arm joint.

[0137] In a further embodiment, the robot is configured that the classification takes into account 3D surveying data generated by the surveying sensor in the upright measurement mode as at least part of the perception data.

[0138] In some embodiments the robot comprises a visual sensor configured to provide at least part of the perception data.

[0139] In some embodiments the robot is configured such that that the crouching measurement mode provides reduced mechanical vibration than the upright measurement mode.

[0140] In some embodiments the surveying criteria define at least one of (i) a desired measurement point density, (ii) a desired coordinate measurement accuracy, (iii) a desired measurement point pattern, (iv) a desired surveying field of view, and (v) a desired perspective onto the object or environment area.

[0141] In some specific embodiment the robot is configured to provide a command for the metrology sensor to alter a resolution and/or sampling rate and/or integration time. In particular wherein the robot is configured to provide said command via a human-machine interface of the metrology sensor. For example the hand joint arrangement of the robot might be sufficiently similar to a human hand that the robot can activate a button, switch or operate a touch screen.

[0142] In some embodiments the robot is switch an operation mode of the surveying sensor. In particular wherein the operation mode affects at least one of an emitted point pattern, an emitted point density, and a coordinate measurement accuracy.

BRIEF DESCRIPTION OF THE DRAWINGS

[0143] By way of example only, specific embodiments of the invention will be described more fully hereinafter with reference to the accompanying figures, wherein:

[0144] Figure 1 depicts schematically an exemplary mobile robot with a track drive.

[0145] Figure 2 depicts schematically a stand up sequence with an inverted state of the two lower parts.

[0146] Figures 3a to 3c depict schematically an embodiment of a quick-release mechanism and its operation.

[0147] Figure 4a depicts schematically a first embodiment of an inventive track-drive pretensioning arrangement. 27

[0148] Figure 4b depicts schematically a second embodiment of an inventive track-drive pretensioning arrangement.

[0149] Figure 5 depicts schematically a ball lock rachet.

[0150] Figure 6 depicts schematically an embodiment of upper leg joint arrangements with two-axes pivotability.

[0151] Figures 7a to 7d depict schematically a mobile robot having a double-track arrangement in different modii.

[0152] Figure 8 depicts schematically to two embodiments for the drive layout for the double track arrangement.

[0153] Figure 9 schematically depicts an embodiment, wherein the leg comprises two drive shafts and a transmission element such that one of the drive shafts provides a mechanical actuation of the other drive shaft via the mechanical actuation element;

[0154] Figure 10 schematically depicts an embodiment, wherein the robot is configured to carry out an iterative tactile sampling of an object to be measured to generate a 3D model of the object;

[0155] Figure 11 schematically depicts a further embodiment of a robot configured to carry out an iterative tactile sampling;

[0156] Figure 12 schematically depicts an embodiment, wherein the robot comprises a depth sensor to acquire 3D images of a mobile scanning device guided by the robot itself, the object to be measured, and a laser tracker that tracks the mobile scanning device, wherein the robot determines and sets a next position and orientation of the mobile scanning device based on an evaluation of the 3D images regarding achievement of a measurement criterion for measuring with the mobile scanning device;

[0157] Figure 13 schematically depicts a working principle of a further embodiment, wherein the robot is configured to be controlled via general text input and to use a closed loop configuration with a large language model (LLM) for robot teaching and interaction of manipulation tasks;

[0158] Figure 14 schematically depicts options to provide the confidence value used by the embodiment described by Fig. 13;

[0159] Figure 15 schematically depicts a working principle of a further embodiment, wherein the robot is configured to learn a surveying task based on input by a gesture of a human operator;

[0160] Figure 16 schematically depicts an exemplary workflow carried out by a further embodiment, wherein the robot is configured to autonomously survey an environment in an upright measurement mode and a crouching measurement mode. SPECIFIC EMBODIMENTS OF THE INVENTION

[0161] Figure 1 depicts an exemplary humanoid robot 1 a, 1 b. While certain aspects of the invention can be realized by a more generic mobile robot the features will be generally illustrated by a humanoid robot. Humanoid in the sense of the invention means that the design of the robot broadly resembles the human body. Particularly the robot comprises a main body 2 configured to pivotably accommodate a set of extremities 3,4 by corresponding pivotable joints 6,11. Said extremities 3,4 might be permanently attached, however they can also be mounted in a releasable manner. For example, while not strictly necessary for some aspects of the invention, said extremities 3,4 are preferably articulated, more preferably multi-articulated.

[0162] The inventive robot following the humanoid template comprises a pair of legs 3 as extremities. Legs 3 in the sense of the invention are extremities configured to provide locomotion of the robot. Even if the preceding sentence is not intended to be interpreted to one of the two extreme cases - i.e., that locomotion is performed solely by motion of the legs 3 and/or the legs 3 cannot perform any action other than locomotion - it is preferable that such locomotion unit is mechanically isolated from the rest of the robot e.g., from the high-precision measuring devices. According to the humanoid template the number of legs 3 is preferably exactly two. While a generally humanlike appearance of the legs 3, as depicted in Figure 1 , is preferable, robots with a plurality of wheels, track drives 31 or comprising one or more adjustable support element per leg 3 are also considered humanoid in the sense of the invention. A robot with legs 3 consisting of a track drive 31 mounted by a respective pivotable joint 6 might also be consider humanoid in the sense of the invention. The legs 3 are attached pivotably to the main body 2. In other words, the humanoid robot is configured to perform a plurality of postures by swiveling the legs 3 with respect to the main body 2. Preferably, the different postures represent different functionalities of the robot. More preferably at least one of the postures is configured for stability, and at least one posture is configured for the speed of locomotion. In preferred embodiments, the humanoid robot is configured for a stepwise motion similar to the walking or running motion of a human.

[0163] The inventive robots following the humanoid template also comprise an arm 4 as extremity. The arm 4 in the sense of the invention is an extremity configured carry out an interaction with the environment, e.g. a contacting or contactless measurement of an object, and/or to grip or otherwise support a tool - e.g., a measuring or manufacturing device - and/or a workpiece - e.g., an object to be examined or processed by the robot -. While advantageous, it is not necessary, that the robot comprise more than one arm. It can be preferable that the robot possesses exactly two multi-articulated arms.

[0164] While not strictly necessary for some aspects of the invention it is preferable that the dimensions and proportions of the robot correspond to a human. More preferably the weight of the robot also corresponds to a weight of human. The advantage is that such a robot can use tools that were developed for human workers and can carry out similar process steps that are currently performed by humans. In addition, training the robot using gesture control is simplified due to the similar proportions and the robot is better suited for human-robot interactions in the manufacturing environment. Additionally, the robot can access the areas of the manufacturing area generally accessible to human workers.

[0165] The robot might also comprise a headlike 5 assembly, particularly comprising a sensor suite and/or communication. Aesthetic reasons aside placing sensing and communication elements on the top of the robot can provide further benefits in the form of a better field of view or signal coverage.

[0166] On the left panel the humanoid robot is in stand-up mode 1 a, while the right panel depicts the humanoid robot in a full track mode 1 b. The robot 1 a, 1 b comprises legs 3 which are attached to the main body 2 by the upper leg or hip joints 6. For transparency reasons some reference signs relating to further identical components, e.g., the second leg 3, are omitted. Each of the legs 3 comprises an upper part 7. In the depicted embodiment the upper part 7 comprises a battery compartment 20. The legs 3 comprise respective lower parts 9. The lower 9 and upper parts 7 are connected to each other by the lower leg or knee joint 8. The knee joint 8 provides a swiveling movement of the lower part 9 relative to the upper part 7. The lower part 9 comprises a track 32 running over a lower pulley 33 and an upper pulley 34. The lower pulley 33 is arranged on the distal end away from the knee joint 8. The upper pulley 34 is arranged closer to the knee joint 8, e.g., wherein its suspension arrangement might comprise elements of the knee joint 8. By way of example, as shown in the figure, the complete lower part 9 is embodied as a track drive 31 .

[0167] The depicted lower part 9 of the leg (the track 32) further comprises one or more support pulleys 35 arranged between the lower pulley 33 and the upper pulley 34 such that the one or more support pulleys 35 is/are in contact with the track 32. For example, the upper pulley 34 is a driven pulley, wherein the upper pulley 34 is driven by utilizing components arranged in the knee joint 8 and/or in the upper part 7 of the leg. The lower pulley 33 might be a driven pulley too, e.g., wherein separate motors drive the upper pulley 34 and the lower pulley 33, respectively. For example, a driven pulley comprises an electric motor integrated into or onto the pulley.

[0168] In the stand-up mode 1 a, a surface contact face of the running surface has smaller area size than a surface contact face of the running surface provided by the full track mode 1 b. In the example, shown, the robot supports itself by standing on a curved part of the track 36 which is curved by the circumferential area of the lower pulley 33. In the stand-up mode 1 a, in particular the locomotion in the stand-up mode 1 a, is provided by a self-balancing over the lower pulley 33. By way of example, the robot comprises an inertial sensor unit 38 in the main body 2, e.g., acceleration sensors and gyroscopic sensors, and a control algorithm configured to automatically control movement of the lower pulleys 33 based on the sensors readings such that the robot is balanced in a defined upright position. [0169] For example, in the full track-mode 1 b the robot supports itself by standing on the largest possible surface contact face provided by the track 37 and the locomotion is provided by track locomotion. In other words, at least a part of the running surface 37 engages the ground. The track drive 31 might comprise a further inertial sensor 39, e.g., an accelerometer. For example, the further inertial sensor 39 in the track drive 31 provides wheelspin, track slippage, or shock information. Said full-track mode 1 b can be advantageously utilized to provide a stable platform, e.g. in a crouching mode as described above.

[0170] The robot is configured to provide the stand-up mode 1 a by a raising of the upper pulleys 32 relative to the lower pulleys 33 to provide selection between two different running surface contact faces, wherein the two contact faces differ from each other by their respective area size. In other words, the robot is configured to provide different "feet sizes" by raising the upper pulleys 34 to different heights above ground and being able to maintain different raised positions.

[0171] In some embodiments the robot is configured to provide track locomotion in the standup mode 1 a. For example, the robot is configured to vary running surface contact faces (vary area sizes) either actively or passively, e.g., by comprising and using additional “displaceable pulleys” (at least one per track drive 31) and a spring pre-forcing to get a pre-forced track. By way of example, the robot can adapt the “feet size” depending on the task of the robot, e.g., adapt the raised position and use a small feet size when standing on position and keeping position is important but use a flatter (larger feet size) for walking up a steep and difficult terrain. Alternatively, the robot might provide a locomotion in the stand-up mode 1 a by stepped motion of the legs 3. During such stepped motion the tracks 32 or a part of the pulleys 33-35 might be immobilized.

[0172] In the depicted embodiment the main body 2 defines a reference system of the robot. In the depicted example the main axes are defined by a leg suspension axis 142 connecting the upper leg or hip joints 6, and a central body axis 141 extending perpendicularly to the leg suspension axis 142. By way of example a central body axis 141 can be foreseen as a kinematic link equivalent to the main body 2. A forward direction 143 and a backward direction 144 can be defined on the basis of the central axis 141 and the leg suspension axis 142.

[0173] The robot also comprises two arms 4 attached to the main body 2 at an upper part of the main body 2 and configured to move relative to the main body 2. Each of the two arms 4 is attached to the main body 2 by a shoulder joint 1 1 , e.g., wherein the shoulder joint 1 1 provides movement in two rotational degrees of freedom relative to the main body. In the example shown, each of the two arms 4 further comprises an elbow joint 12 and a hand joint arrangement 13. Each of the elbow joints 12 provides movement in one rotational degree of freedom, e.g., for providing an arm folding movement. The depicted hand joint arrangement 13 is configured to provide for performing a gripping operation. By way of example, the hand joint arrangement 13 is configured to provide a hand embodied like a human hand. Alternatively, the hand joint arrangement provides for a simpler gripping tool like a two or three fingered claw. In another alternative embodiments the hand joint arrangement 13 might comprise metrology grade sensors.

[0174] The robot might be further configured to provide the self-balancing movement by the aid of an automatic movement control for movement of the arms 4 during locomotion so that movements by the arms 4 support the robot being balanced in the defined upright position. By way of example, the robot comprises a control algorithm configured to automatically adjust the relative pose of the two mechanically actuated arms based on the tilt sensors and the gyroscopic sensors so that the weight and orientation of the arms 4 provide a stabilizing effect to maintain the upright position of the robot 1 .

[0175] The robot also comprises a head 5 and arranged to a head joint 15 connecting the head 5 to the main body 2. The head joint 15 might provide at least one degrees of freedom movability of the head 5 with respect to the main body 2. The head 5 might comprise perception, such as a lidar unit 23, one or more time of flight camera 26, one or more high field of view camera 27 or one or more microphone 30. The head 5 might also comprise a positioning unit 29 providing positioning data based on external positioning signals, such as GNSS, WLAN or Bluetooth signals. Alternatively, or additionally the main body 2 could comprise a part of, or all the above sensors. Said sensors 23, 26, 27, 29, 30 provide data, in addition or alternatively of the inertial sensor unit 38, for the control algorithm configured to automatically control movement of the lower pulleys 33 based on the sensors readings such that the robot is balanced in a defined upright position.

[0176] Mounting the sensors to one of the limbs, wherein the head 5 is also comprised by the limbs, is advantageous as such arrangement (i) could enable a quick exchange of the sensor block, and (ii) mounts the sensors far away from the locomotion unit, i.e., provides a low vibration environment of the sensors. Nonetheless many aspects of the claimed invention can be realized by robots comprising neither arms 4 nor head 5 as additional extremities.

[0177] Embodiments wherein the sensors are distributed to a plurality of limbs provide the advantage of different viewpoints, e.g., (i) the robot might assess (lower-resolution) perception data regarding the environment using a perception sensor mounted in the head and metrology data using a sensor mounted on the hand joint arrangement, (ii) the robot might position one of the hand joint arrangements such that it has a view of first face of an object and the position the another hand joint arrangement such it has a view of a second face, different from the first face, or (iii) the robot might use its extremities to provide "a base" with variable base geometry e.g., for triangulation measurements. The above list is non-exhaustive, and the skilled person could provide suitable alternatives in the sense of the invention.

[0178] In some embodiments the robot comprises exactly two mechanically actuated legs 3, exactly two mechanically actuated arms 4 attached to an upper portion of the main body 2 and configured to move relative to the main body and a, particularly mechanically actuated, head 5 attached to the at the top of the main body 2, as depicted. [0179] Preferably each of the arms 4 is attached to the main body by an upper arm joint providing movement 11 in one rotational degree of freedom, particularly two rotational degrees of freedom, relative to the main body 2. Preferably each of the arms 4 comprises a lower arm joint 12 providing movement in one rotational degree of freedom and a hand joint arrangement 13 configured for performing a gripping operation.

[0180] While not explicitly shown, the humanoid robot is preferably configured to provide the locomotion in a walking mode by stepped motion of the legs 3, in particular wherein the running surfaces 37 are immobilized in the walking mode.

[0181] While not explicitly shown preferably the robot is configured to maintain - during the locomotion - an orientation of the main body 2 with respect to a defined posture relative to the gravity direction, particularly wherein at the same time an orientation of the running surface 37 relative to the ground is maintained.

[0182] While not explicitly shown preferably the running surface 37 of the track comprises a structure, in particular a microstructure, configured to increase a static friction coefficient of the running surface 37. In particular wherein the static and dynamic friction coefficient of the structured running surface 37 is anisotropic.

[0183] While not explicitly shown the robot might be is configured such that each of the lower leg joints 8 is driven by a respective electric motor located away from the lower leg joint 8. Particularly each of the legs 3 comprises a mechanical transmission element driven by the respective electric motor, in particular a belt or a chain, to provide mechanical actuation of the movement about the respective lower leg joint 8.

[0184] While not explicitly shown the robot might comprise a ground characterization sensor configured to provide a ground quality information, in particular a parameter providing wheelspin information. The robot might be configured to automatically set different surface contact faces of the running surface 37 on the basis of the ground quality information, particularly by providing a raising and lowering of the upper pulley 34 to different raised positions above ground and selfbalancing in each of the different raised positions. The ground characterization sensor might comprise a set of inertial sensors configured to provide (i) track slippage information, wherein track slippage represents a relative motion to the ground of a part of the track engaging the ground, and (ii) shock information, wherein a shock represents an acceleration, in particular in the vertical direction, above a threshold.

[0185] It goes without saying that whenever an embodiment of a humanoid robot is referred to below, the robot may include one or more of the features described above.

[0186] Figure 2 depicts a second embodiment of the inventive humanoid robot during a standup sequence. Initially the robot is the full track mode 1 b, such that a first vector 147r extending from an upper pulley 34r to a lower pulley 33r of a first 31 r of two track drives 311,31 r and a second vector 1471 extending from an upper pulley 341 to a lower pulley 331 of a second 311 of the two track drives 311,31 r both points to the forward direction 143. By swiveling 1461 a second lower leg joint 81 connecting the corresponding lower and upper parts an inverted state is achieved, wherein the second vector 1471 points backwards 144, i.e., the opposite of the first vector 147r. In the depicted embodiment such inverted state is realized such that each of the lower pulleys 33l,33r is farther away from the main body 2 than the corresponding upper pulley 34l,34r.

[0187] The depicted inverted state might be seen as a part of a stand-up sequence 1 c or an alternative embodiment of the full track mode 1 b. When it is part of the stand-up sequence 1 c it can follow, as depicted, by raising the upper pulleys 34l,34l from the ground by swiveling 1461,146r the lower leg 8r,8l joints in opposing directions. Advantageously, the lower pulleys 33l,33r are far away from the main body 2, thus the depicted stand-up sequence 1 c is more resistant to the inevitable small instabilities.

[0188] By way of example the respective upper parts 7l,7r of the leg are shown to be static during the stand-up sequence 1 c. This embodiment was selected for transparency reasons only, actual embodiments of the stand-up sequence

[0189] By way of example depicted robot comprises two different hand joint arrangements 131,13r. One 13r of them is depicted to resemble a human hand, while the other 131 comprises a metrology suit with a visual sensor 21 , a metrology grade one-point tactile sensor 22 and a touch sensor 148. The depicted hand arrangements 131,13r are combinable with further embodiments or can be replaced by suitable alternatives, particularly from other depicted embodiments.

[0190] Figures 3a to 3b schematically depict an embodiment of the extremity locking mechanism 161. By way of example the depicted extremity locking mechanism 161 contacts the main body 2 with one of the legs 3 (only schematically shown). The extremity locking mechanism 161 is provided by a receptacle 150 having a depression for receiving a spigot 149 along a penetration axis 152. By way of example the receptacle 150 is arranged to the main body 2. This arrangement is prudent as the spigot 149, which is more susceptible to damage, is arranged to the lower-cost part, nevertheless the reverse arrangement is also possible within the meaning of the invention.

[0191] The receptacle 150 and the spigot 149 have matching cornered, particularly hexagonal, shapes. One of the at least three, particularly six, latching elements 151 of the spigot is shown. Said latching element 151 is depicted as a ball. Alternative embodiments of the latching element 151 , such as an ellipsoid, a trapezoid, a pyramid, a trapezoid having rounded corners, or a pyramid having rounded corners etc. are also possible within the sense of the invention.

[0192] Figure 3b depicts a basic position of a release mechanism 153b arranged in the spigot 149. In said basic position 153b each of the latching elements 151 pushes radially outwards from the penetration axis 152 in order to engage in a corresponding cavity 154 of the receptacle 150. In the depicted embodiment the latching element 151 has at least two points of contact with the corresponding cavity 154.

[0193] By way of example the latching elements 151 and the cavities 154 are configured and matched to each other in such a manner that the engagement of the latching elements 151 in the cavities 154 causes a self-centering of the spigot 149, particularly with respect to the penetration axis 152. Optionally the engagement of the latching elements 151 in the cavities 154 further causes a self-alignment of corresponding rotational angles in a plane perpendicular to the penetration axis 152. Preferably the self-centering and self-alignment provides a fixed initial angle for the leg 3. The latching elements 151 might be mounted in a non-rotational symmetric manner to the spigot 149. Alternatively the receptacle 150 and the spigot 149 have shapes that only allow inserting the spigot 149 into the receptacle 150 in a single pre-defined relative orientation with respect to each other.

[0194] Figure 3a depicts an activated position of the release mechanism 153a. The activation of the release mechanism 153a enables the latching elements 151 to radially escape into the spigot 149 in order to allow the spigot 149 to be released from the receptacle 150.

[0195] The depicted release mechanism 153a, 153b comprises two radial pins 158 for activating the release mechanism 153a, 153b. Alternative embodiments with only one or more than two radial pins 158 are also possible in the sense of the invention. The release mechanism 153a, 153b also comprises an axial pin 160 for blocking or allowing radial escape of the latching elements 151 and a tensioning spring 159 to maintain the basic position 153b. The radial pins 158, the axial pin 160, and the tensioning spring 159 are operatively connected in such a way that (i) in the basic position of the release mechanism 153b the axial pin 160 forces the latching elements 151 radially outwards, and (ii) when the release mechanism is activated a displacement of the radial pins 158 moves the axial pin 160 towards the tensioning spring 159, and the axial pin 160 releases space due to its displacement and thus enables the radial escape of the latching elements 151 into the spigot 149. In the depicted basic position 153b the axial pin 160, by means of a tensioning force due to the tensioning spring 159, presses the latching elements 151 into the cavities 154.

[0196] Figure 3c illustrates a manner by which in a mounted state of the leg 3, i.e., in the basic position of the release mechanism 153b, a flow of electrical energy is enabled between the leg 3 and the main body 2. For that the extremity locking mechanism 161 comprises a set of pogo-pins 155 arranged on a front face of the spigot 149 and a set of corresponding recesses 157 arranged on a corresponding face of the depression in the receptable 150. The pogo-pins 155 and the recesses 157 are configured to engage each other thereby providing an electrical connection between the leg 3 and the main body 2. The integrated helical spring 255 in the pin 155 applies a constant normal force against the recess 157. Thus, such pogo-pins 155 are especially suited for providing stable electrical connection between moving components. [0197] The extremity locking mechanism 161 also comprises a dismounting functionality configured to block the flow of electrical energy between the leg 3 and the main body 2. This can be performed e.g., by a switch 162 in operative contact by the radial pins and arranged such that it terminates the electric feed lines 156 in the leg 3 or in the main body 2.

[0198] Figure 4a depicts a track drive 31 comprising a first embodiment of the inventive pretensioning arrangement 163 configured to provide tensioning of the track 32 by providing a tensioning force acting to increase a distance between the lower pulley 33 and the upper pulley 34. The depicted pretensioning arrangement 163 is configured to set the tensioning force by making use of an impact force 164 acting on the pretensioning arrangement 163 in the stand-up position. Alternatively, the pretensioning arrangement 163 might use the weight of the robot for the same purpose.

[0199] The pretensioning arrangement 163 comprises a spring-based tensioning unit with a lower fixed component 165, an upper fixed component 166, and a floating component 167. The lower fixed component 165, marked with shading, has a fixed spatial relationship with a suspension of the lower pulley 33. The upper fixed component 166, marked with dotting, has a fixed spatial relationship with a suspension of the upper pulley 34. The floating component 167 is arranged between the upper 34 and lower pulley 33, and a relative position of the floating component 167 and one of the upper 166 or lower fixed components 165 defines the tensioning state of the springbased tensioning unit.

[0200] The depicted pretensioning arrangement 163 also comprises a counter mechanism. In the depicted embodiment the counter mechanism is provided by a latch arrangement 168 connecting the floating component 167 with the upper 166 and lower fixed components 165. The depicted latch arrangement 168 comprises a first latch component 170 between the floating body 167 and the upper fixed component 166. The first latch component 170 is configured to restrict a movement of the floating body 167 with respect to the upper fixed component 166 to a first unidirectional movement 172. The latch arrangement 168 also comprises a second latch component 169 between the floating body 167 and the lower fixed component 165. The second latch component 169 is configured to restrict a movement of the floating body 167 with respect to the lower fixed component 165 to a second unidirectional movement 171 . The first 172 and second unidirectional movement 171 have opposite direction. Some embodiments of the latch arrangement 169, in particular ball lock ratchets (an exemplary embodiment is shown on Figure 5), are configured to provide a stepless tensioning of the track. Other embodiments of the latch arrangement, in particular linear ratches, are configured to provide incremental steps to tension the track.

[0201] The depicted floating body 167 comprises an axial member 173 arranged along an axis 174 connecting the upper 34 and lower pulleys 33, such that is axially displaceable along said axis 174. The depicted floating body 167 also comprises a peripherical member 175. The peripherical member 175 is arranged transversally offset from said axis, and it is transversally displaceable 177 by the axial displacement 176 of the axial member 173. In the depicted embodiment the axial 173 and the peripherical members 174 has a conical or frustrum shapes and two peripherical members 175 are arranged symmetrically to the axial member 173.

[0202] The depicted pretensioning arrangement causes portions of the track 32 located between the upper 34 and lower pulleys 33 to align to a common tangent of said pulleys 33,34. In other words, the pretensioning arrangement does not comprise deflection pulleys to tension the tracks.

[0203] For transparency reasons, the track drives 31 are shown in a vertical position. This position is advantageous as the weight of the robot can be optimally utilized to provide the impact force. The claimed invention is however not limited to these cases.

[0204] Figure 4b depicts schematically a second embodiment of the depicted pretensioning arrangement 163 comprising a lower fixed component 165, an upper fixed component 166, and a floating component 167. The lower fixed component 165, marked with shading, has a fixed spatial relationship with a suspension of the lower pulley 33. The upper fixed component 166, marked with dotting, has a fixed spatial relationship with a suspension of the upper pulley 34.

[0205] The spring 178 connects the floating body 167 and the lower fixed component 165 such that an impact force 164 causes a displacement of the floating body 167. The spring 178 is a high stiffness spring, wherein high stiffness in the sense of the invention means that the spring experience no significant deformation owing to the weight of the robot and/or the weight of the track drive 31 . In other words, the spatial relations of the members of the track drive 31 is invariant irrespectively whether track drive is horizontal, vertically hanging on the upper pulley 34 or vertically supporting the weight of the robot. The impact force 164 might be provided by pre-programmed sequence of stepping, particularly a trampling movement.

[0206] The pretensioning arrangement 163 also comprises a latch arrangement connecting the floating component 167 with the upper 166 and lower fixed components 165. The latch arrangement comprises a first latch component 170 between the floating body 167 and the upper fixed component 166. The first latch component 170 is configured to restrict a movement of the floating body 167 with respect to the upper fixed component 166 to a first unidirectional movement 172. The latch arrangement also comprises a second latch component 169 between the floating body 167 and the lower fixed component 165. The second latch component 169 is configured to restrict a movement of the floating body 167 with respect to the lower fixed component 165 to a second unidirectional movement 171 . The first 172 and second unidirectional movement 171 have opposite direction.

[0207] A first spring connects 179 the floating body 167 and the upper fixed component 166. A rope 180, connects an anchoring point 184, in the proximity to first spring 179 and transmits a force to the lower fixed component 165. The upper fixed component 166 has a rope guide surface 181 , depicted as a surface with a semicircular cross-section. The rope guide surface 181 acts as a fixed deflection pulley with respect to the rope 180 and thereby converts the direction of pull 182 of the rope 180. The rope 180 is anchored 183 at the lower fixed component 165 and provides second unidirectional movement 171 via its pull 182.

[0208] While not depicted in the figures the utilization of alternative embodiment of the components, particularly hydraulic components, can be utilized alternatively or additionally to depicted ones.

[0209] Figure 5 schematically depicts a ball lock rachet 300 configured to provide a stepless tensioning of the track. The ball lock rachet 300 comprises a cavity 304 between a first body 302 (e.g., the lower fixed component) and a second body 303 (e.g., the floating component). By way of example the first body 302 is depicted as a bulk object, while the second body 303 is depicted as a thin sheet. The skilled person understands that this selection is purely arbitrary, both of the first 302 and second bodies 303 can be equally realized as bulk bodies and/or sheets. Moreover the cavity 304 might extend into the envelope volume of the first body 302. A ball 305, or a suitable alternative object, is disposed within the cavity 304. The cavity is nonsymmetrical and comprises a wide "enabling-position" 306 and a narrow "blocking-position" 308.

[0210] By a relative movement of the first body 302 to the second body 303 in the enabling direction 301 a the ball 305 is conveyed to the enabling-position 306. In the enabling-position 306, the ball 305 can be easily dislocated by a displacement or rotation 307, so that it offers negligible resistance to movement in the enabling direction 301 a.

[0211] By a relative movement of the first body 302 to the second body 303 in the blocking direction 301 b the ball 305 is conveyed to the blocking-position 308. In the blocking position 308 the ball is pressed 309,310 against both the first 302 and the second bodies 303, and thereby hinders a movement in the blocking direction 301 b.

[0212] Figure 6 depicts an embodiment of upper leg joint arrangements 6 with two-axes pivotability. Said joint arrangement 6 comprises a first part 185, closer to the leg 3, providing a pivotability about a first swivel axis 187. Each of the legs 3 can be swiveled independently of the other leg about a respective first swivel axis 187 of the corresponding upper leg joint arrangement 6. The first swivel axes 187 and an axis 142 connecting the upper leg joint arrangements 6 are located in the same plane.

[0213] The upper leg joint arrangement 6 comprises a second part 186. The second part 186 is configured to provide a pivotability for the corresponding leg 3 about a second swivel axis 188 of the upper leg joint arrangement 6. The second swivel axis 188 is substantially orthogonal to the axis 142 connecting the upper leg joint arrangements 6, respectively the first swivel axis 187.

[0214] For reasons of transparency the two swivel axes 187,188 of the depicted upper leg joint arrangement 6 are realized by two functionally and spatially distinct components 185,186. This serves merely to ensure transparency and readability and the present invention can be realized by a single multidimensional joint, e.g., similarly to a human hip. The specific features of such embodiments might be applied accordingly.

[0215] A respective motor unit 189 for providing the pivotability for the corresponding leg 3 about the second swivel axis 188 is arranged offset from the leg joint arrangement 6. The depicted motor units 189 are arranged in the main body 2. It is clear for the skilled person that inventive leg joint arrangement 6 can also be realized such that the respective motor units 189 are disposed with the legs 3. The specific features of such embodiments might be applied accordingly.

[0216] The depicted motor unit 189 is embodied as a linear actuator with (i) an axle 190 aligned to a direction substantially orthogonal to the first 187 and second swivel axis 188, (ii) a traveler 191 configured to be linearly displaceable along the axle 190, and (iii) a motor 192 to move the traveler 191 along the axle 190.

[0217] The traveler 191 is kinematically linked by a transfer element 193 to a rotary element 194 of the second part 186 of the upper leg joint arrangement 6 for providing the pivotability of the leg 3 about the second swivel axis 188. The transfer element 193 and the rotary element 194 are configured to translate a linear movement 195 of the traveler 192 to a rotational movement 196 of the rotary element 194. In other words, a linear movement 195 of the traveler 191 causes a swiveling of the leg 3 about the second swiveling axis 188. By way of example the transfer element 193 is depicted as a rigid link, however the present invention is not limited to such embodiments. The transfer element might equally be realized by a rope or a ball-screw drive. The second part 186 might comprise a counter element configured to provide a restoring force acting to restore a default position of the rotary element 194.

[0218] The depicted motor unit 189 comprises a ball-screw gear 197. A ball-screw gear 197 advantageously allows a compact realization of the motor unit 189 by allowing a remote actuation of the traveler 191 .In the depicted embodiment the axle 190 is static and the motor directly drives the ball screw gear 197. In alternative embodiments according to the invention the motor 192 drives the axle 190, particularly when the ball screw gear 197 is mounted to the traveler 191 .

[0219] In spite of said benefits of the ball screw gear 197 other embodiments, particularly, wherein the motor 192 and the traveler 191 are realized as a single-piece construct, are also possible in the sense of the invention.

[0220] The depicted motor unit 189 comprises a plurality passive damping elements 198, embodied as a set of springs, arranged to an end point of the axle 190. The damping elements 198 are configured to provide a restore force acting on the traveler 191 and a causing a departure of the traveler 191 from the endpoint.

[0221] In the depicted embodiment the first part 185 of the upper leg joint arrangement 6 comprises a further rotary element 199 for providing a swiveling of the respective leg 3 about the respective first swivel axis 187. Said further rotary element 199 is located outside a volume defined by a housing of the main body 2, in the particular embodiment is an integral part of the leg 3. The further rotary element 199 might be realized as an electric motor. The first part 185 of the upper leg joint arrangement 6 might also be realized in a similar manner to the second part 186.

[0222] The traveler might comprise a plurality of parts, in the depicted embodiment a lower 202 and an upper part 203 of the traveler, separated by a respective bearing 204. Advantageously the upper part 203 might rotate about the axle 190 during the linear movement 195 of the traveler 191 , while the lower part 202 maintain its relative orientation to the rotary element 194.

[0223] Figure 7a depicts a mobile robot 101 a comprising a main body 2 and two mechanically actuated legs attached to the main body 2 at a lower portion 205 of the main body 2 by respective upper leg 6 joints (only one is shown). Each of the legs 3l,3r comprises an upper part 7 and a corresponding set of track drives 31 ,208 attached to the upper part 7 at a corresponding lower leg joint 8.

[0224] Each of the track drives 31 ,208 comprising a track running over respective lower and upper pulleys and providing a running surface for track locomotion. The upper pulleys 34,209 of the first 31 and second track drives 208 of the first leg 3r are mounted coaxially to the corresponding lower leg joint 8 of the first leg 3r, such that the first 34 and second track drives 209 of the first leg 3r are independently pivotable about the axis of the lower leg joint 8 as a track swivel axis of the first leg 3r. The construction of the second leg 3I is analogous. By way of example the depicted the second track drives 31 disposed on the side of the upper part of the leg 7 opposite of the first track drives 208. Alternatively or additionally, the second track drives 31 might be stacked on the first track drives 208.

[0225] The depicted robot 101 a comprises two articulated arms 4l,4r and a perception sensor suite comprising a stereo camera 40. Due to the number of extremities said mobile robot 101 a can also be understood as humanoid. Alternative embodiments with more arms 4l,4r are also possible in the sense of the present invention. By way of example one of the arms 4I comprises a locking port 206 configured to provide a fixed mechanical connection with the environment, while the other arm 4r comprises an integrated metrology sensor 207, e.g., a lidar.

[0226] Figure 7b depicts the same robot in an extended track mode 101 b. In the extended track mode 101 b for each of the legs the respective first 31 and second track drives 208 are twisted 211 to each other about the respective track swivel axis so that the upper pulleys 34,209 are located between the lower pulleys 33,210. In the extended track mode track locomotion is provided by running surfaces of all the first 31 and second track drives 208, particularly wherein the upper and lower pulleys of the first and second track drives are arranged to a common plane. The extended track drive mode is especially advantageous as it provides a longer running surface, i.e., a more stable platform especially on uneven ground. Moreover, the robot can utilize the extended track drive mode for ditch crossing and stair climbing. An exemplary realization of stair climbing is shown in Figure 7c. This also implies that the length each of the first and second track exceeds a typical distance between two stairs, in particular it exceeds 40 cm.

[0227] Figure 7c depicts the mobile robot in a climbing mode 101 c. In the climbing mode 101 c an angle between a first track vector extending from the upper pulley 34 of the first track drive 31 to the lower pulley 33 of the first track drive 31 and a second track vector extending from the upper pulley 209 of the second track drive 208 to the lower pulley 210 second track drive 208 falls into a range of 15° to 55. Said angle corresponds to the typical ascent angle 213 of stairs 212, i.e., the mobile robot is configured to foreseen to climb common stairs.

[0228] An exemplary stair climbing sequence can be foreseen as (i) identifying a stair 212 (e.g., by the stereo camera 40), (ii) setting the angle 215 between the track vectors 147,214 to correspond to the ascent angle 213 of the stair 212, (iii) approach the stair such that first track drives 208 establish contact with the stair, (iv) optionally adapt the orientation of the main body 2, i.e., by swiveling the track drives 31 ,208 about the track swivel axis and/or the main body 2 about the upper leg joint 6; and (v) drive over the stair 212 using the first track drives 208 and continuously adapting the angle 215 between the first 208 and second track drives 31. In other words, when contact with the stair is established both the first and second track drives maintain at least two- point contact with the ground.

[0229] Figure 7d an alternative stand-up sequence to the one depicted in Figure 1 c based on the extended track mode 101 b. Said extended track stand-up sequence 101d can be performed by raising the upper pulleys 34,209 from the ground by swiveling each of first 31 and second track drives 208 in a coordinated manner such that the robot 101d supports itself on the four lower pulleys 33,210. Optionally the orientation of the main body 2 might also be adapted.

[0230] It is clear for the skilled person that suitable alternatives of the above sequence exist, thus the above sequence shall not be construed in a limiting manner.

[0231] Especially for the depicted climbing mode 101 c it is advantageous when the running surfaces of the track drives 31 ,208 are structured.

[0232] Figure 8 depicts, with dashed lines, an exemplary embodiment of a leg 3 as it can be used in a robot according to the invention. The leg comprises an upper part 7 and lower part 9 such that upper 7 and lower parts 9 are connected by a knee joint 8 configured to provide a swiveling movement of the lower part 9 relative to the upper part. The upper part 7 of the leg comprises a drive shaft 46 mechanically connected to an electric motor 47. The knee joint comprises a further shaft 48 mechanically connected to the knee joint, such that a rotation and/or swiveling movement of the further shaft 48 causes a swiveling movement of the knee joint. The drive shaft 46 and the further shaft 48 are connected by a mechanical transmission element 49 such that the drive shaft 46 provides a mechanical actuation of the further shaft 48 via the mechanical transmission element 49. In the depicted embodiment the drive shaft 46 and the further shaft 48 each comprise a toothed area and the mechanical transmission element 49 is a chain drive. Alternative embodiments might comprise (non-exhaustive list) a belt drive, e.g., a v-belt, or a toothed bar as mechanical transmission element 49.

[0233] The depicted arrangement might be used to drive the swiveling of one ofthe track drives at the lower part 7 of the leg 3 as well as to provide mechanical energy to the driven pulleys of track drives.

[0234] Figure 9 depicts schematically the drive arrangement of a mobile robot having one first 208, and two second track drives 31 a, 31 b. The first track drive 208 and one of the second track drives 31 a are located on an outer face of the upper part 7 of the leg, wherein the outer face is the face more remote from the main body 2 (only the interface is shown). The other second track drive 31 b is located at the opposite side.

[0235] The upper parts 7 of the leg comprises two coupling interfaces 216 configured to releasably attach the respective first track drive 208 and the respective second track drive 31 b to opposite sides of the respective leg. The first track drive 208 comprises an analogous coupling interface 217 configured releasably attach the respective second track drive 31 a. In the depicted embodiment the coupling interfaces 216 of the upper part 7 of the leg and the coupling interface 217 of the first track drive 208 are identical thus the second track drives 31 a, 31 b could be attached to any of them. This also means that the first track drives 208 are also stackable to each other.

[0236] In alternative embodiments the coupling interfaces 216,217 might be different, i.e., one type might be configured to releasable attach only the first 208 or the second track drives 31 a, 31 b. In another alternative embodiments the first 208 and second track drives 31 a, 31 b are essentially identical and only the stacking order differentiates them from other. Matching counterparts 218,219 of a coupling interface 216,217 are respectively comprised by the first 208, and second track drives 31 a, 31 b.

[0237] The first track drive 208-and the second track drives 31 a, 31 b are attached so that they can rotate around the respective track swivel axis, here the axis 220 of the lower leg joint 8, such that the first 208 and second track drives 31 a, 31 b are independently pivotable about said track swivel axis 220.

[0238] In the depicted embodiment the pivotability of the track drives 31 a, 31 b, 208 is provided by a set of rotating motor components 223 arranged in a position spaced away from the lower leg joint 8. The lower leg joint 8 comprises a set of shafts 48 mechanically connected to the coupling interfaces 216 such that swiveling movement of the further shaft 48 causes a swiveling movement of the coupling interfaces 216. The set of rotating motor components 223 and the set of shafts 48 are connected by respective mechanical transmission elements 49 such that the rotating motor components 223 provides a mechanical actuation of the shafts 48 via the mechanical transmission elements 49. In alternative embodiments the rotating motor components are arranged to the track drives 31 a, 31 b, 208 or along the track swivel axis 220 in the upper part 7 of the leg. [0239] In the depicted embodiment an additional interface part 221 is passed through the first track drive 208 into the leg axially to the track swivel axis 220. At a distal end inside the upper part 7 of the leg the additional interface part 221 is configured to engage a rotating motor component

223 inside the upper part 7 of the leg. The rotational movement about the track swivel axis 220 is transmitted from the rotating motor component 223 the second track 31 a drive by means of the additional interface part 221 . In the depicted embodiment the additional interface part 221 engages the coupling interface 216 of the upper part 7 of the leg and comprises a gear arrangement 222 to provide the independent pivotability from the first track drive 208.

[0240] In the depicted embodiment the track drives 31 a, 31 b, 208 comprise respective track drive motors 224 configured to mechanically actuate the respective driven pulleys. Said motors might be mounted directly to the pulleys. Alternative embodiments, wherein the track drive motors

224 are located outside the respective track drives 31 a, 31 b, 208 is also possible.

[0241] The skilled person understands that many of the depicted features are optional. Particularly embodiments wherein the sole first track drive 208 and the sole second track drives 31 a are mounted only at outer face of the upper part 7 of the leg or alternatively wherein the sole first track drive 208 and the sole second track drives 31 b are mounted at opposite face of the upper part 7 of the leg are within the sense of the invention. Stacking further first 208 and second track drives 31 a, 31 b is also possible. Moreover, with corresponding adjustments, the features of the depicted embodiments are transferable wherein the first 208 and second track drives 31 a are directly mounted to the main body 2, e.g., at the upper leg joint.

[0242] Figures 10 and 1 1 schematically depict a humanoid robot according to embodiments of the invention, wherein the robot is configured to carry out an iterative tactile sampling of an object to be measured to generate a 3D model of the object. The tactile sampling is carried out by a touch sensor 450, 451 , wherein in the embodiment depicted by Fig. 10, the robot is configured to hold an external touch sensor 450 in its hand whereas in the embodiment depicted by Fig. 11 , the robot itself comprises the touch sensor 451 , e.g. wherein the touch sensor is arranged in a hand of the robot. The touch sensor 450, 451 comprises an elastomeric component for contacting a surface patch 452 of the object to be measured 453, wherein the touch sensor 450, 451 is configured to generate tactile sensing data based on observation of deformation of an inner face of the elastomeric component when an outer face of the elastomeric component is brought into physical contact with the object to be measured. Such a touch sensor thus provides for measuring an extended surface patch 452 of the object in one go.

[0243] For example, such a touch sensor 450, 451 often comprises a reflective coating on the inner face of the elastomeric component, wherein touching of an object with the outer face of the elastomeric component causes a reproduction of a geometric shape of a portion of the touched object on the inner face. The 3D information of the object 453 is then derived by imaging this reproduction of the geometric shape of the object on the reflective inner face of the elastomeric component by an imaging arrangement disposed within the touch sensor interior volume.

[0244] The iterative tactile sampling comprises movement of the arm and the hand of the robot to change poses (position and orientation) of the touch sensor 450, 451 to provide for touching different surface patches 452 with the elastomeric component. In order to be able to generate the 3D model of the object 453, the robot is further configured to generate pose data that provide for referencing tactile sensing data associated with different poses of the touch sensor 450, 451 to a common coordinate system.

[0245] By way of example, the pose data comprise kinematic chain data determined by angular encoders of the robot for tracking movement of the hand and the touch sensor 450, 451. Alternatively or in addition, the pose of touch sensor is tracked by one or more visual sensors of the robot, e.g. a camera-based or light-based tracking sensor such as a stereo arrangement of cameras or a TOF camera. For example, such a visual tracking sensor is arranged in the main body or in the head of the robot and, for example, comprises an active illumination source (e.g. for providing a flash or continuous light). As a further option, a point cloud matching procedure is used to provide referencing between tactile sensing data of the touch sensor associated with (overlapping) surface patches measured by the touch sensor.

[0246] In a particular embodiment, the robot comprises a visual sensor 454, 455 configured to generate visual data of different surface areas of the object, and the robot is configured to provide for referencing the visual data relative to the tactile sensing data. For example, the visual sensor 454, 455 is embodied as at least one of a high-resolution camera, a time-of-flight camera (TOF camera, often also referred to as range imaging camera, RIM camera), a LIDAR sensor, and a structured light scanner. In the embodiment depicted by Fig. 10, the visual sensor 454 is arranged in the main body of the robot. In the embodiment depicted by Fig. 1 1 , the visual sensor 455 is embodied as a low focal length camera arranged on the arm of the robot in close proximity to the touch sensor 451 , e.g., wherein a distance measured along the arm to a hand-joint arrangement for holding the sensor is two times shorter, more particularly ten times shorter, than a distance to the main body.

[0247] The visual data are used to estimate an optimized pose of the touch sensor 450, 451 during the iterative tactile sampling. Visual data often complement tactile data of the touch sensor, e.g. wherein the visual data provide object texture (color) and visual data are often more accurate than the sparse tactile point-cloud at edges where touch sensors 450, 451 are often not as accurate. Thus, the visual data may also be used for surface reconstruction of the object by photogrammetric principles (e.g. structure from motion, SFM) where tactile sensing is not that accurate. For example, tactile sensing is good for flat and smooth surfaces but may become less accurate in case of steeper steps in the object surface. [0248] By way of example, the visual data are used to determine a confidence value for a measurement with the touch sensor 450, 451 , e.g. by identifying flat surfaces and/or contrast changes exceeding a contrast change threshold, and, based thereof, to optimize placement of the touch sensor 450, 451 on the object 453 during the iterative tactile sampling.

[0249] Figure 12 schematically depicts a humanoid robot according to an embodiment of the invention, wherein the robot is configured to hold and position a metrology grade mobile scanning device 500 in a fixed position and orientation relative to an object to be measured 501 . By way of example, the mobile scanning device 500 is configured for scanning the object 501 by means of laser-based scanning and/or camera-based scanning. The mobile scanning device comprises multiple cooperative targets 502, e.g. retro-reflectors, for being tracked by a stationary laser tracker 503. In addition, the mobile scanning device 500 comprises markings 504 that allow determination of the 6DoF pose of the scanning device 500. For example, the laser tracker 503 comprises a 6DoF-camera for detecting and analyzing the markings 504 by image processing.

[0250] The robot is configured to autonomously move the mobile scanning device 500 to different positions and orientations relative to the object 501 and, based thereof, to use the mobile scanning device 500 to generate 3D position measuring data of the object 501 according to a defined measurement criterion. The robot comprises a depth sensor 505, e.g. a TOF camera, to acquire 3D images of the mobile scanning device 500, the object to be measured 501 , and the laser tracker 503 in order to determine and set a next position and orientation of the mobile scanning device 500 based on an evaluation regarding achievement of the measurement criterion.

[0251] By way of example, the measurement criterion is at least one of a point density, a distance measurement accuracy, a coordinate measurement accuracy, and a signal-to-noise ratio to be achieved by the measurement with the mobile scanning device 500. Based on the 3D information of the object 501 and previous relative positions and orientations between the mobile scanning device 500, the object 501 , and the laser tracker 503, determined from data of the depth sensor 505, the evaluation provides an optimization of a scanning distance and/or a viewing angle of the mobile scanning device 500 to the object to be measured 501. In other words, as both the positions of the object to scan 500 and the laser tracker 503 can be estimated (tracked) by the humanoid robot visually by the depth camera 505 (and optionally further built-in 2D or 3D cameras), an optimized pose of the mobile scanning device 500 can be calculated, e.g. by also using tracking data and/or 6DoF pose determination by the laser tracker 503.

[0252] For example, the robot is further configured to determine track data for tracking position and orientation of the mobile scanning device 500 relative to a position of the laser tracker 503 and the robot by using the distance measuring data of the depth sensor 505 and to use the track data to identify and avoid upcoming line-of-sight breaking between the laser tracker 503 and the mobile scanner 500. [0253] Figure 13 schematically depicts a working principle of a further embodiment of a humanoid robot according to the invention, wherein the robot is configured to be controlled via general text input and to use a closed loop configuration with a large language model (LLM) for robot teaching and interaction of manipulation tasks, e.g. in combination with locomotion tasks.

[0254] As part of a perception sensor, the robot comprises a visual perception sensor 506, e.g. a 2D or 3D camera, and is configured to use a metrology-grade surveying sensor. For example, the metrology-grade surveying sensor is part of the robot or is held by a hand-joint arrangement of the robot arranged at a distal end of an arm of the robot. By way of example, the metrology-grade surveying sensor is configured to provide for point- or line-measurements, e.g. based on stereoimaging or other kind of photogrammetry such as structured light scanning.

[0255] The robot further comprises a large language model module (LLM-module) 507 for processing text input of a planned surveying task to be executed by the robot, wherein the planned surveying task involves a specific movement of the robot arm relative to the object. The LLM- module is trained 508 to transform the text input of the planned surveying task to an output comprising robot command language. By way of example, the LLM-module is implemented by means of a machine learning algorithm, e.g. an artificial neural network (pre-)trained using selfsupervised learning and/or semi-supervised learning. The output is analyzed by a robot arm instruction interpreter 509, configured to select robot commands out of the output of the LLM- module 507. For example, the complete text of the LLM-module output is parsed and only parts in robot command language are selected, e.g. parts in JSON or XML. A robot arm instruction executor 510 of the robot then causes the robot to execute the robot commands selected by the robot arm instruction interpreter 509, resulting in an executed surveying task executed by the robot.

[0256] By way of example, the robot comprises a speech to text module 511 for transforming spoken instructions into text instructions provided to the LLM-module 507. Thanks to pre-training 508 and conditioning of the LLM-module 507, the output of the LLM-module is providing general text including specific robot commands, e.g. an instruction list in parse-able form. The robot may further be configured to provide feedback regarding the output of the LLM-module to a user, e.g. in text form via a display 512 or in audible form via a text to speech module 513.

[0257] The robot further comprises a comparison module 514 configured to provide a comparison of the planned surveying task and the executed surveying task based on perception data of the perception sensor 506 to provide a confidence value regarding achievement of the planned surveying task, wherein the robot is configured to take into account the confidence value to adapt movement of the robot when repeating the planned surveying task. The perception data capture a relative positional relationship between the object and at least part of the robot, e.g. the arm, and thereby provide visual feedback of the executed surveying task.

[0258] By way of example, the closed loop for providing the confidence value may further include - similar to the robot arm instruction interpreter 509 and the robot arm instruction executor 510 - a locomotion instruction interpreter 515 and a locomotion instruction executor 516. The locomotion instruction interpreter 515 selects the specific robot commands for locomotion out of the general text output by the LLM-module 507, and forwards them to locomotion instruction executor 516, which then executes the instruction for locomotion of the robot and moves the robot to the instructed place (physical action). For example, locomotion sensors 517 comprising at least one of an IMU (inertial measurement unit), a camera, and LiDAR (depth sensing) take notion of the robot locomotion and provide input for a simultaneous localization and mapping (SLAM) module 518. The comparison module 514 then calculates the distance between the locomotion instruction and the actual position reached and provides feedback to the LLM-module 507.

[0259] For example, the robot is configured to automatically recognize whether the planned surveying task is correctly executed or whether additional robot instructions are required. Thus, task interpretation by the robot is incrementally improved, e.g., wherein the robot is configured to stop operation as a function of a counter of successively derived confidence values below or above a defined threshold.

[0260] By way of example, the comparison module 514 is configured to provide the confidence value by analyzing the text input of the planned surveying task to provide a planned position of the robot relative to the object to be measured and to compare the planned position and an actual position of the robot relative to the object.

[0261] Other options to provide the confidence value are schematically depicted by Figure 14. The comparison module compares a parameter space 519 associated to a current state of the robot 520 and/or the environment 521 with a defined allowable parameter space 522 associated with an allowed state of the robot 523 and/or the environment 524. The perception data to determine a current pose or movement pattern may include visual tracking data capturing parts of the robot and the environment, e.g. by 2D or 3D camera data or laser scanning data. In addition, the robot may be configured to track position and orientation of the arm and the surveying device based on position data of a kinematic chain of moving parts of the robot, e.g. by using data of angle encoders in the arm and/or legs of the robot.

[0262] By way of example, the actual state of the robot is defined by movement sensors 525 of the robot, e.g. providing encoder and IMU readings as well as 2D image information 526 and 3D information 527, e.g. provided by the visual perception sensor 506 and possibly further cameras of the robot. This information is compared to a pre-defined list of values 528 for the movement sensors 525 and allowed relative positions 529 of robot parts relative to other robot parts (robot selfcapture). The actual state of the environment is determined by 2D image information 530 of the environment and 3D information of the environment 531 , e.g. provided by the visual perception sensor 506 and possibly further cameras of the robot. The state of the environment is then compared to a defined range 532 of allowable locations of an object described in the task, e.g. to determine whether any change of the object is as expected. The state of the environment may also be compared to a capture 533 of the environment before and after the task, e.g. to determine whether something else than the object to be measured changed and whether this change is as expected.

[0263] For example, the confidence value is provided by determining an actual pose of at least part of the robot and comparing the actual pose with a list of known poses of the at least part of the robot and a corresponding defined range of a permissible pose deviation for each of the known poses. Alternatively or in addition, the confidence value is provided by determining an actual movement pattern of the robot and comparing the actual movement pattern with defined permissible movement patterns of the robot. A further way to provide the confidence value is by determining a change in an environment of the robot before and after execution of the executed surveying task. For example, the robot is configured to recognize a position of part of the robot relative to the object to be measured and a change of positions of further objects in the environment.

[0264] Known (allowable) poses of the robot may be defined by physical constraints to provide metrology measurements with sufficient accuracy and repeatability. For example, certain arm positions provide unwanted increased vibration on the surveying sensor, e.g. when the arm is stretched far away from the main body of the robot. The robot may have a defined set of preferred stable poses and movement patterns in an upright or crouching robot position for providing surveying data by the surveying sensor with sufficient quality. It may also be defined that the robot is not allowed to fall below certain minimal distances to objects of the environment. On the other hand, optimal measurement distances and viewing angles with defined tolerance levels may be associated to different types of measurements using the metrology grade surveying sensor, and the robot may not be allowed to exceed these tolerance levels.

[0265] Figure 15 schematically depicts a working principle of a further embodiment of a humanoid robot according to the invention, wherein the robot is configured to learn a surveying task based on input by a gesture 534 of a human operator. The surveying task involves specific movements of the robot and the use of a metrology-grade surveying sensor. The metrology-grade surveying sensor is part of the robot or is held by the robot (see also above).

[0266] The robot comprises a perception sensor 535 to generate visual perception data and a gesture recognition module 536 making use of the visual perception data. The gesture recognition module is configured to analyze the perception data to recognize a gesture 534 made by an operator that is associated with a planned surveying task. The gesture by the operator is translated into a planned model movement sequence 537 of the robot. An instruction interpreter 538 provides robot commands based on the planned model movement sequence 537 and a robot instruction executor 539 is configured to cause the robot to execute the robot commands, resulting in an executed surveying task 540 executed by the robot. The robot further comprises a confirmation module 541 configured to analyze perception data of the perception sensor 535 that capture a movement of the robot when executing the robot commands. The observed movement is translated into an executed model movement sequence 542, which is compared to the planned model movement sequence 537 in order to provide a confidence value 543 regarding a matching of the planned and executed model movement sequences.

[0267] In other words, the robot is configured to observe a gesture 534 by a human operator and to copy the gesture, wherein the robot carries out a self-observation to compare the observed movement 534 by the human operator with its own movement 540 that should - within physical and/or defined constraints 544 of the freedom of movement of the robot - be a copy of the observed movement 534 by the human operator. The confidence value 543 provides indication of a matching of the planned movement model sequence 537 and the executed movement model sequence 542. If the movement model 537 derived from observing the human operator and the movement model 542 derived from observing the robot itself do not match, the robot carries out an adapted executed surveying task when repeating the planned surveying task.

[0268] By way of example, the confidence value 543 is taken into account to adapt processing of the gesture recognition module 536 and/or to adapt processing of the instruction interpreter 538.

[0269] In particular, the confidence value 543 may also be provided based on measurement data of the metrology-grade surveying sensor and/or measurement data of the perception sensor (and/or other sensors of the robot) capturing the environment of the robot. For example, such data are used to recognize and take into account a size difference between the human operator and the robot when deriving the robot commands for the planned movement model sequence 537. Based on matching size (and matching constraints on the freedom of movement of the robot compared to a human) the planned model movement sequence 537 is provided to resemble a one-to-one copy of the human movement or the planned model movement sequence 537 is provided to resemble a scaled sequence such that the same effect is generated (e.g. a same measurement distance to the object to be measured) when measuring with the metrology-grade surveying sensor.

[0270] Figure 16 schematically depicts an exemplary workflow carried out by a further embodiment of a humanoid robot according to the invention, wherein each of the legs of the robot comprises a track drive providing a track for track locomotion and the robot is configured to execute surveying tasks by a metrology-grade surveying sensor 500 in an upright measurement mode 545 and a crouching measurement mode 546. The surveying sensor 500 may be embodied by an external hand-held surveying device, wherein the robot is configured to hold the device in its hand. Alternatively, the surveying sensor 500 may be an integral sensor of the robot.

[0271] In the upright measurement mode 545, the robot "stands" on its legs such that the main body of the robot is placed higher above ground than in the crouching measurement mode 546 and a posture of the robot has to be balanced based on a control algorithm configured to automatically stabilize the posture based on inertial data of an inertial sensor of the robot (e.g. comprising at least one of an accelerometer and a gyroscope). In the crouching measurement mode 546, the robot resumes a tilt-resistant posture by "kneeling" such that the robot is supported on an extended track area of its track drives and the tilt-resistant posture can be maintained free of active balancing by the robot.

[0272] The robot is further configured to capture perception data of the environment, e.g. by means of a visual perception sensor 506, and to carry out a classification of objects and environment areas within the perception data based on object type and area type. By way of example, the classification is carried out by means of a neural network or other kind of machine learning algorithm. According to this aspect, different object types and different area types are assigned different surveying criteria for surveying with the surveying sensor 500 and the robot is configured to automatically switch between the upright measurement mode 545 and the crouching measurement mode 546 based on the classification.

[0273] By way of example, this provides for a workflow wherein the robot is provided with a task 547 involving the capture of the environment with the surveying sensor 500 (and possibly other sensors). For example, the robot is given the task to autonomously survey a defined perimeter within a building. The robot then starts surveying the environment with the metrologygrade surveying sensor 500 in the upright measurement mode 545 to generate a first set of 3D data 548 of the environment, wherein the upright measurement mode 545 allows faster surveying than the crouching measurement mode 546. During surveying in the upright measurement mode 545 the robot generates a map 549 of the environment by using perception data of its perception sensor 506 (and possibly also data of the metrology-grade surveying sensor 500) and identifies areas and objects to be measured by the crouching measurement mode 546, which provides surveying of these areas and objects with different measurement configurations and/or measurement conditions, e.g. wherein the crouching measurement mode 546 provides increased stability and thus increase precision of the surveying sensor 500 or different viewing angles and distances onto surfaces to be measured. This information is then used as feedback to autonomously change to the crouching measurement mode 546 for surveying these areas and objects and thereby generating a second set of 3D data 550 of the environment. In the end, the first set of 3D data 548 and the second set of 3D data 550 is combined to provide a complete capture 551 of the environment and to derive a digital 3D model of the environment.

[0274] For example, the robot moves through the complete environment in the upright measurement mode 545 and generates a map of the environment 549 with stored locations to be followed-up by the crouching measurement mode 546 in a second step. Alternatively, the robot frequently changes between the upright measurement mode 545 and the crouching measurement mode 546, e.g. each time when an area or object to be surveyed in the crouching measurement mode 546 is identified.

[0275] By way of example, the crouching measurement mode 546 includes the robot to resume a well-defined stable holding position of the surveying sensor 500. In particular, the surveying sensor 500 that is held in the hand of the robot or the arm of the robot itself is locked to a locking interface at the main body of the robot. This provides a stable and vibration reduced surveying pose of the surveying sensor 500, e.g. wherein the robot-surveying sensor combination can essentially act like a stationary surveying apparatus. [0276] In particular, the robot is configured to generate the 3D surveying data 548, 550 with motion correction based on compensation data of an inertial measurement unit. However, when resuming an intrinsically stable position in the crouching measurement mode 546, the motion compensation algorithm - which may require minimal motion to work properly - can generate increased noise. Therefore, a weighting of the compensation data for the motion correction is reduced in the crouching measurement mode 546 compared to the upright measurement mode 545. In particular, motion compensation is completely deactivated in the crouching measurement mode 546 and a vibration-free, and particularly a motion-free, pose of the surveying sensor 500 is assumed for generation of the 3D surveying data 550 in the crouching measurement mode 546.

[0277] Although the invention is illustrated above, partly with reference to some specific embodiments, it must be understood that numerous modifications and combinations of different features of the embodiments can be made. All of these modifications lie within the scope of the appended claims.

Claims

1 . Humanoid robot, comprising a main body and two mechanically actuated legs attached to the main body at a lower portion of the main body by respective upper leg joints located at opposing sides of the main body, wherein

• each of the legs can be swiveled independently of the other leg about a respective swivel axis of the corresponding upper leg joint,

• each of the legs comprises corresponding upper and lower parts, which are connected to each other via a lower leg joint and can be swiveled against each other about an axis of the lower leg joint, and

• each of the lower parts comprises a track drive comprising a track running over a lower pulley and an upper pulley, thereby providing a running surface between the lower and the upper pulley, wherein the lower pulley is arranged at a distal end away from the lower leg joint and the upper pulley is arranged closer to the lower leg joint than the lower pulley, characterized in that each of the lower leg joints is configured to provide 360° rotatability for the lower part about the axis of the lower leg joint and the robot is configured to provide a stand-up sequence, comprising

• establishing an inverted state of the two lower parts, wherein a first vector extending from the upper pulley to the lower pulley of one of the two lower parts and a second vector extending from the upper pulley to the lower pulley of the other of the two lower parts point to opposite sides of the robot, particularly wherein the first vector points to an area in front of the robot and the second vector points to an area in the back of the robot, and

• raising the upper pulleys from the ground by swiveling the lower leg joints in opposing directions.

2. The humanoid robot of claim 1 , wherein in the inverted state, the first vector comprises a forward pointing component and the second vector comprises a backward pointing component, defined with respect to o a central body axis disposed between the upper leg joints and extending in a direction perpendicular to a leg suspension axis connecting the upper leg joints, and o a middle plane comprising the leg suspension axis and the central body axis, wherein the forward and backward pointing components are perpendicular to the middle plane and point to opposite directions.

3. The humanoid robot according to claim 2, wherein in the inverted state each of the lower pulleys is farther away from the central body axis than the corresponding upper pulley.

4. The humanoid robot according to claim 3, wherein in the inverted state one of the lower parts is completely located in a front domain and the other lower part is completely located in a rear domain, wherein the front domain is the domain on one side of the middle plane and the rear domain is the domain on the other side of the middle plane.

5. The humanoid robot according to claim 2, wherein in the inverted state each of the lower parts is partly located in a front domain and partly located in a rear domain, wherein the front domain is the domain on one side of the middle plane and the rear domain is the domain on the other side of the middle plane, in particular wherein one upper pulley and one lower pulley are located in the front domain and one upper pulley and one lower pulley are located in the rear domain.

6. The humanoid robot according to any one of the preceding claims, wherein the axes of the lower leg joints are parallel to each other.

7. The humanoid robot according to any one of the preceding claims being configured to provide locomotion by track locomotion both in a stand-up mode and in a full track mode, wherein the stand-up mode provides a surface contact face of the running surface that has smaller area size than a surface contact face of the running surface provided by the full track mode, which is achieved by the robot automatically arranging the lower part relative to the upper part such that during the locomotion the upper pulley is raised to a raised position that is farther from ground than a position of the lower pulley.

8. The humanoid robot according to claim 7, wherein during the stand-up sequence the robot supports itself by self-balancing locomotion to maintain the raised position of the upper pulley, in particular wherein the stand-up sequence ends in a stand-up position where the robot supports itself by solely standing on a curved part of the track which is curved by the circumferential area of the lower pulley.

9. Humanoid robot according to claim 8 comprising an inertial sensor, particularly embodied as a set of accelerometers and the gyroscopes, and a control algorithm configured to automatically control the track drive based on the inertial sensor such that the robot is balanced in a defined upright position associated with the raised position of the upper pulley.

10. Humanoid robot, comprising a main body and mechanically actuated extremities attached to the main body, wherein the mechanically actuated extremities comprise two mechanically actuated legs attached to the main body at a lower portion of the main body by respective upper leg joints, wherein each of the legs can be swiveled independently of the other leg about a swivel axis of the corresponding upper leg joint, and the legs are configured to provide locomotion of the robot over ground, characterized in that at least one of the mechanically actuated extremities is releasably attachable to the main body by an extremity locking mechanism, wherein the extremity locking mechanism is provided by a receptacle having a depression for receiving a spigot along a penetration axis, wherein

• the receptacle and the spigot have matching cornered, particularly hexagonal, shapes,

• the spigot has at least three, particularly six, latching elements, wherein in a basic position of a release mechanism, each of the latching elements pushes radially outwards in order to engage in a corresponding cavity of the receptacle, and

• an activation of the release mechanism enables the latching elements to radially escape into the spigot in order to allow the spigot to be released from the receptacle.

11 . The humanoid robot according to claim 10, wherein the robot is configured that it comprises the teaching of one of claims 1 to 9.

12. The humanoid robot according to any one of claims 10 to 11 , wherein each of the latching elements is configured as a rotation body, in particular as a sphere or an ellipsoid, a trapezoid, a pyramid, a trapezoid having rounded corners, or a pyramid having rounded corners.

13. The humanoid robot according to any one of claims 10 to 12, wherein the latching elements and the cavities are configured and matched to each other in such a manner that the engagement of the latching elements in the cavities causes a self-centering of the spigot, in particular a self-centering with respect to the penetration axis.

14. The humanoid robot according to any one of claims 10 to 13, wherein the release mechanism is arranged in the spigot and has

• at least one radial pin for activating the release mechanism,

• an axial pin for blocking or allowing radial escape of the latching elements, and

• a tensioning spring to maintain the basic position, wherein the radial pin, the axial pin and the tensioning spring are operatively connected in such a way that • in the basic position of the release mechanism the axial pin forces the latching elements radially outwards, and

• when the release mechanism is activated o a displacement of the radial pin moves the axial pin towards the tensioning spring, and o the axial pin releases space due to its displacement and thus enables the radial escape of the latching elements into the spigot.

15. The humanoid robot according to claim 14, wherein in the basic position of the axial pin, by means of a tensioning force due to the tensioning spring, presses the latching elements into the cavities.

16. The humanoid robot according to any one of claims 10 to 15, wherein each of the latching elements has at least two points of contact with its corresponding cavity.

17. The humanoid robot according to any one of claims 10 to 16, wherein the extremity locking mechanism is configured to provide

• a mounted state, wherein the at least one of the mechanically actuated extremities is mechanically constrained by the degrees of freedom provided by a joint attaching the at least one of the mechanically actuated extremities to the main body, and a flow of electrical energy is enabled between the at least one of the mechanically actuated extremities and the main body, and

• a dismounting functionality configured to block the flow of electrical energy between the at least one of the mechanically actuated extremities and the main body and to cancel at least a part of the mechanical constraints between the at least one of the mechanically actuated extremities and the main body, thereby allowing a separation of the at least one of the mechanically actuated extremities from the main body.

18. The humanoid robot according to claim 17, wherein the extremity locking mechanism comprises a set of pogo-pins arranged on a front face of the spigot and a set of corresponding recesses arranged on a corresponding face of the depression in the receptable, wherein the pogo-pins and the recesses are configured to engage each other thereby providing an electrical connection between the mechanically actuated extremity and the main body.

19. The humanoid robot according to any one of claims 10 to 18, wherein the receptacle and the spigot have shapes that only allow inserting the spigot into the receptacle in a single pre-defined relative orientation with respect to each other, in particular, wherein the engagement of the latching elements in the cavities causes a self-alignment of corresponding rotational angles in a plane perpendicular to the penetration axis, more particular wherein the self-centering and self-alignment provides a fixed initial angle for the leg.

20. Humanoid robot, comprising a main body and two mechanically actuated legs attached to the main body at a lower portion of the main body by respective upper leg joints located at opposing sides of the main body, wherein

• each of the legs can be swiveled independently of the other leg about a respective swivel axis provided by the corresponding upper leg joint,

• each of the legs comprises corresponding upper and lower parts, which are connected to each other via a lower leg joint and can be swiveled against each other about an axis provided by the lower leg joint,

• each of the lower parts comprises a track drive comprising a track running over a lower pulley and an upper pulley, thereby providing a running surface between the lower and the upper pulley, wherein the lower pulley is arranged at a distal end away from the lower leg joint and the upper pulley is arranged closer to the lower leg joint than the lower pulley and the robot is configured to angle the lower parts relative to the upper parts to support the robot on the running surfaces to provide locomotion by the track drive, and

• the robot is configured to provide a stand-up position, wherein the lower parts are raised relative to the upper parts such that the upper pulleys are raised to raised positions that are farther from ground than positions of the lower pulleys, particularly wherein the running surfaces are vertical, characterized in that each of the track drives comprises a pretensioning arrangement configured to provide tensioning of the track by providing a tensioning force acting to increase a distance between the lower pulley and the upper pulley, wherein the pretensioning arrangement is configured to set the tensioning force by making use of a gravitational force and/or an impact force acting on the pretensioning arrangement in the stand-up position.

21 . The humanoid robot according to claim 20, wherein the robot is configured that it comprises the teaching of one of claims 1 to 19.

22. The humanoid robot according to any one of claims 20 to 21 , wherein each of the pretensioning arrangements comprises a counter mechanism configured to set a tensioning state of a spring-based tensioning unit in reaction to the gravitational force and/or the impact force, in particular wherein the tensioning state of the spring-based tensioning unit is provided in incremental steps as a function of a tension of the track.

23. The humanoid robot according to claim 22, wherein the spring-based tensioning unit comprises an upper fixed component, a lower fixed component, and a floating component, wherein

• the floating component is arranged between the upper and lower pulley, the upper fixed component is fixed with the upper pulley, and the lower fixed component is fixed with the lower pulley,

• the counter mechanism is provided by a latch arrangement connecting the floating component with the upper and lower fixed components, such that movement of the floating component relative to the upper fixed component and movement of the floating component relative to the lower fixed component are restricted to different unidirectional directions of movement, and

• a relative position of the floating component and one of the upper or lower fixed components defines the tensioning state of the spring-based tensioning unit.

24. The humanoid robot according to claim 23, wherein

• the spring-based tensioning unit comprises two springs,

• a first of the two springs connects the floating body and the upper fixed component and a second of the two springs connects the floating body unit and the lower fixed component,

• the latch arrangement comprises a first latch component between the floating body and the upper fixed component configured to restrict a movement of the floating body with respect to the upper fixed component to a first unidirectional movement,

• the latch arrangement comprises a second latch component between the floating body and the lower fixed component configured to restrict a movement of the floating body with respect to the lower fixed component to a second unidirectional movement, wherein the first and second unidirectional movement have opposite directions.

25. The humanoid robot according to claim 24, wherein

• the pretensioning arrangement comprises a rope connecting the second of the two springs and the lower fixed component and transmitting a spring force of the second of the two springs to the lower fixed component,

• the upper fixed component has a rope guide surface, in particular in the form of a surface with a semicircular cross-section, which acts as a fixed deflection pulley with respect to the rope and thereby converts the direction of pull of the rope.

26. The humanoid robot according to any of claims 24 to 25, wherein the second spring is high stiffness spring.

27. The humanoid robot according to any one of claims 20 to 26, wherein the pretensioning arrangement comprises

• an axial member arranged along an axis connecting the upper and lower pulleys, such that is axially displaceable along said axis, and

• a peripherical member o arranged transversally offset from said axis, such that it is transversally displaceable by the axial displacement of the axial member, and o being configured to set by its transversal displacement a distance between the upper and lower fixed components, in particular wherein the axial member has a conical or frustrum shape, and the peripherical member has a conical or frustrum shape, and/or in particular wherein the floating body comprises a further peripherical member arranged symmetrically to the peripherical member.

28. The humanoid robot according to any one of the claims 20 to 27, wherein the tensioning of the tracks causes portions of the track located between the upper and lower pulleys to align to a common tangent of said pulleys.

29. Humanoid robot, comprising a main body and two mechanically actuated legs attached to the main body at a lower portion of the main body by respective upper leg joint arrangements located at opposing sides of the main body, wherein

• each of the legs can be swiveled independently of the other leg about a respective first swivel axis of the corresponding upper leg joint arrangement for deflecting the legs forwards and backwards for providing a walking motion of the robot, particularly wherein the first swivel axes and an axis connecting the upper leg joint arrangements are located in the same plane, and

• each of the legs comprises a track drive comprising a track providing a running surface for track locomotion, characterized in that each of the upper leg joint arrangements is configured to provide a pivotability for the corresponding leg about a second swivel axis of the upper leg joint arrangement that is different from the respective first swivel axis, particularly wherein the second swivel axis is orthogonal to the axis connecting the upper leg joint arrangements, wherein a respective motor unit for providing the pivotability for the corresponding leg about the second swivel axis is arranged offset from the leg joint arrangement, wherein the motor unit comprises a linear actuator, particularly arranged in the main body, comprising

• an axle,

• a traveler configured to be linearly displaceable along the axle, and

• a motor to move the traveler along the axle, wherein the traveler is kinematically linked by a transfer element to a rotary element of the upper leg joint arrangement for providing the pivotability of the leg about the second swivel axis, wherein the transfer element and the rotary element are configured to translate a linear movement of the traveler to a rotational movement of the rotary element.

30. The humanoid robot according to claim 29, wherein the robot is configured that it comprises the teaching of one of claims 1 to 28.

31 . The humanoid robot according to any one of claims 29 to 30, wherein the humanoid robot is configured to provide a walking locomotion by shifting its center of gravity by a coordinated swiveling of the legs about the corresponding second swivel axis.

32. The humanoid robot according to any one of claims 29 to 31 , wherein each of the linear actuators comprises at least one passive damping element, particularly embodied as a spring, arranged to an endpoint of the axle and configured to provide a restore force acting on the traveler and a causing a departure of the traveler from the endpoint.

33. The humanoid robot according to any one of claims 29 to 32, wherein the linear actuator comprises a ball-screw gear.

34. The humanoid robot according to any one of claims 29 to 33, wherein a displacement range of each of the second swivel axes is at least 10°, in particular wherein each of the second swivel axes are configured to provide an inward tilt of the corresponding legs.

35. The humanoid robot according to any one of claims 29 to 34, wherein each of the leg joint arrangements comprises a further rotary element for providing a swiveling of the respective leg about the respective first swivel axis, wherein the further rotary elements are located outside a volume defined by a housing of the main body.

36. The humanoid robot according to any one of claims 29 to 35 being configured to maintain a defined orientation of the main body with respect to a direction of gravity by the coordinated swiveling of the legs about the corresponding second swivel axis.

37. The humanoid robot according to any one of claims 29 to 36, configured to maintain contact between the ground and each of the running surfaces by a coordinated swiveling of the legs about the corresponding second swivel axis.

38. Mobile robot, comprising a main body, two mechanically actuated legs attached to the main body at a lower portion of the main body by respective upper leg joints, characterized in that

• each of the legs comprises respective first and second track drives, each of the track drives comprising a track running over respective lower and upper pulleys and providing a running surface for track locomotion,

• the upper pulleys of the first and second track drives of the first leg are mounted coaxially to a track swivel axis of the first leg, such that the first and second track drives of the first leg are independently pivotable about the track swivel axis of the first leg, and

• the upper pulleys of the first and second track drives of the second leg are mounted coaxially to a track swivel axis of the second leg, such that the first and second track drives of the second leg are independently pivotable about the track swivel axis of the second leg.

39. The mobile robot, in particular humanoid robot, according to claim 38, wherein the robot is configured that it comprises the teaching of one of claims 1 to 37.

40. The mobile robot according to any one of claims 38 to 39, wherein

• each of the legs comprises corresponding upper and lower parts, which are connected to each other via a lower leg joint such that they can be swiveled against each other,

• the respective first and second track drives are comprised by the lower parts of the corresponding leg, and

• the respective track swivel axes are provided by the corresponding lower leg joints.

41 . The mobile robot according to claim 40, wherein the first and second track drives are located on an outer face of the upper part of the leg, wherein the outer face is the face more remote from the main body.

42. The mobile robot according to any one of claims 38 to 41 configured to provide an extended track mode, wherein in the extended track mode for each of the legs the respective first and second track drives are twisted to each other about the respective track swivel axis so that the upper pulleys are located between the lower pulleys, wherein in the extended track mode track locomotion is provided by running surfaces of all the first and second track drives, particularly wherein the upper and lower pulleys of the first and second track drives are arranged to a common plane.

43. The mobile robot according to claim 42 being configured to provide an extended track stand-up sequence, comprising

• establishing the extended track mode for both of the legs and

• raising the upper pulleys from the ground by swiveling each of first and second track drives in a coordinated manner such that the robot supports itself on the four lower pulleys.

44. The mobile robot according to any one of claims 38 to 43 being configured to provide a climbing mode, wherein in the climbing mode an angle between a first track vector extending from the upper pulley of the first track drive to the lower pulley of the first track drive and a second track vector extending from the upper pulley of the second track drive to the lower pulley second track drive falls into a range of 15° to 55°.

45. The mobile robot according to claim 44 being configured to adjust the angle between the first and second track vectors such that both the first and second track drives maintain at least two-point contact with the ground.

46. The mobile robot according to any one claims 38 to 45, wherein for each of the legs, the respective leg and at least one of the respective first track drive and the respective second track drive comprise matching counterparts of a coupling interface configured to releasably attach the at least one of the respective first track drive and the respective second track drive to the respective leg, wherein the coupling interface is configured to provide for independent swiveling of the first track drive and the second track drive about the track swivel axis.

47. The mobile robot according to claim 46, wherein the coupling interface is configured to releasably attach the respective first track drive and the respective second track drive to opposite sides of the respective leg.

48. The mobile robot according to any one claims 46 to 47, wherein the coupling interface is configured to releasably attach the respective second track drive to the respective first track drive, so that the respective leg and the respective first and second track drives are stacked one above the other in the order leg-first track drive-second track drive so that they can rotate around the respective track swivel axis of the leg.

49. The mobile robot according to claim 48, wherein

• an additional interface part is passed through the first track drive into the leg axially to the track swivel axis, wherein at a distal end inside the leg the additional interface part is configured to engage a rotating motor component inside the leg so that a rotational movement about the track swivel axis is transmitted from the rotating motor component inside the leg to the second track drive by means of the additional interface part, and

• at a distal end on side of the second track drive the second track drive is attached with its counterpart of the coupling interface to a matching counterpart on the additional interface component, particularly wherein the first track drive and the second track drive comprise identical counterparts of the coupling interface.

50. The mobile robot according to any one claims 46 to 49, wherein

• each of the first and second track drives comprises a corresponding track drive motor configured to provide a driving force for at least one of the respective lower or upper pulleys, and

• each of the legs comprises a first and a second swivel motor, wherein the coupling interface is configured that the first swivel motor drives a swiveling of the first track drive about the track swivel axis and the second swivel motor drives a swiveling of the second track drive about the track swivel axis.

51 . The mobile robot according to claim 50, wherein the coupling interface is configured such that the second swivel motor drives two shafts, wherein the two shafts are arranged and configured to engage coupling interface components that are approached to the leg from opposite sides of the leg.

52. The mobile robot according to any one of claims 46 to 51 , wherein the coupling interface is provided by a receptacle having a depression for receiving a spigot along the track swivel axis, wherein

• the spigot has at least three, particularly six, latching bodies, wherein in a basic position of a release mechanism, each of the latching bodies pushes radially outwards in order to engage in a corresponding cavity of the receptacle, and

• an activation of the release mechanism enables the latching bodies to radially escape into the spigot in order to allow the spigot to be released from the receptacle, particularly wherein the coupling interface is embodied as the extremity locking mechanism according to one of claims 10 to 19.

53. Humanoid robot, comprising a main body, two mechanically actuated legs attached to the main body at a lower portion of the main body by respective upper leg joints, and a mechanically actuated arm attached to the main body by an upper arm joint, wherein the legs and the arm can be swiveled independently of one another, wherein the humanoid robot comprises a depth measuring sensor, configured to generate distance measuring data to an environment of the robot, characterized in that the humanoid robot is configured to use the arm to hold and position a mobile scanning device in a plurality of positions and orientation relative to an object to be measured, wherein the mobile scanning device is configured to approach the object to be measured and to generate 3D position measuring data of the object to be measured, wherein the humanoid robot is configured to autonomously move the mobile scanning device to different positions and orientations relative to the object to be measured and, based thereof, to use the mobile scanning device to generate 3D position measuring data of the object to be measured according to a defined measurement criterion, wherein the humanoid robot is configured to determine and set a next position and orientation of the mobile scanning device by moving the arm based on an evaluation regarding achievement of the measurement criterion taking into account previous positions and orientations of the mobile scanning device, known measurement characteristics of the mobile scanning device, and distance measuring data of the depth measuring sensor for measuring distances to the object to be measured and the mobile scanning device.

54. The humanoid robot according to claim 53, wherein the robot is configured that it comprises the teaching of one of claims 1 to 52.

55. The humanoid robot according to claim 53 or 54, wherein the measurement criterion is at least one of a point density, a distance measurement accuracy, a coordinate measurement accuracy, and a signal-to-noise ratio.

56. The humanoid robot according to any one of claims 53 to 55, wherein the mobile scanning device is embodied as a laser-based scanner, a tactile scanning device, or a stereo imaging device.

57. The humanoid robot according to any one of claims 53 to 56, wherein the depth measuring sensor is embodied as a laser-based distance measuring device, a time-of-flight camera, or a stereo imaging device.

58. The humanoid robot according to any one of claims 53 to 57, wherein the evaluation is configured to provide an optimization of a scanning distance and/or a viewing angle of the mobile scanning device to the object to be measured in order to meet the measurement criterion.

59. The humanoid robot according to any one of claims 53 to 58, configured to determine the previous positions and orientations of the mobile scanning device by tracking position and orientation of the mobile scanning device by the distance measuring data of the depth measuring sensor.

60. The humanoid robot according to any one of claims 53 to 59, configured to determine the previous positions and orientations of the mobile scanning device by tracking position and orientation of the mobile scanning device based on position data of a kinematic chain of moving parts of the robot, in particular the arm and the legs of the robot.

61 . The humanoid robot according to any one of claims 53 to 60, wherein the mobile scanning device is configured to be tracked by a tracking device, particularly an industrial laser tracker, wherein the humanoid robot is configured to determine the next position and orientation of the mobile scanning device based on distance measuring data of the depth measuring sensor for measuring distances to the mobile scanning device and the tracking device, by evaluating relative positions and orientations of the mobile scanning device relative to the object to be measured and the tracking device.

62. The humanoid robot according to claim 61 , wherein the humanoid robot is configured to determine track data for tracking position and orientation of the mobile scanning device relative to a position of the tracking device and the robot by using the distance measuring data of the depth measuring sensor and to use the track data to identify and avoid upcoming line-of-sight breaking between the tracking device and the mobile scanner.

63. The humanoid robot according to any one of claims 53 to 62, wherein the evaluation regarding achievement of the measurement criterion takes into account feedback from the mobile scanner.

64. Humanoid robot, comprising a main body, a mechanically actuated arm attached to the main body, and a hand-joint arrangement at a distal end of the arm remote from the main body, wherein

• the hand-joint arrangement is configured to hold a touch sensor or the hand-joint arrangement comprises a touch sensor, wherein the touch sensor comprises an elastomeric component for contacting a surface patch of an object to be measured and is configured to generate tactile sensing data that provide geometric 3D information of the surface patch based on observation of deformation of an inner face of the elastomeric component when an outer face of the elastomeric component is brought into physical contact with the object to be measured, and

• the robot is configured to carry out an iterative tactile sampling of different surface patches of the object with the elastomeric component, thereby generating tactile sensing data associated with the different surface patches, wherein the iterative tactile sampling comprises movement of the arm and the hand-joint arrangement to change poses of the touch sensor to provide for touching the different surface patches with the elastomeric component, and generation of pose data that provide for referencing tactile sensing data associated with different poses of the touch sensor to a common coordinate system.

65. The humanoid robot according to claim 64, wherein the robot is configured that it comprises the teaching of one of claims 1 to 63.

66. The humanoid robot according to claim 64 or 65, wherein the pose data comprise kinematic chain data determined by angular encoders of the robot for tracking movement of the arm and/or the hand-joint arrangement.

67. The humanoid robot according to any one of claims 64 to 66, wherein the pose data comprise tracking data determined by a camera-based or light-based tracking sensor of the robot configured to capture at least part of at least one of the arm, the hand-joint arrangement, and the tactile sensor, particularly wherein the camera-based or light-based tracking sensor is arranged in the main body or in a mechanically actuated head attached to the main body at the top of the main body, more particularly wherein the camera-based or light-based tracking sensor has an active illumination source included.

68. The humanoid robot according to any one of claims 64 to 67, wherein the robot is configured to provide the pose data by carrying out a point cloud matching procedure on tactile sensing data associated with overlapping surface patches of the different surface patches.

69. The humanoid robot according to any one of claims 64 to 68, wherein the robot is configured to use the tactile sensing data and the pose data to generate a 3D model of the object.

70. The humanoid robot according to any one of claims 64 to 69, wherein the robot comprises a visual sensor configured to generate visual data of the different surface patches, and the robot is configured to provide for referencing the visual data relative to the tactile sensing data, particularly wherein the robot is configured to use the visual data to texture the 3D model of the object.

71 . The humanoid robot according to claim 70, wherein the robot is configured to use the visual data during the iterative tactile sampling to determine a next surface patch to be measured by the touch sensor.

72. The humanoid robot according to claim 71 , wherein the next surface patch is determined based on a surface reconstruction of the object by processing the visual data based on a photogrammetric principle, particularly based on a structure from motion algorithm.

73. The humanoid robot according to any one of claims 70 to 72, wherein the robot is configured to use the visual data to determine a tactile sensing confidence value of the tactile sensing data and

• to adapt placement of the touch sensor on the object during the iterative tactile sampling and/or

• to adapt a data assignment to complement the tactile sensing data with the visual data based on the tactile sensing confidence value.

74. The humanoid robot according to claim 73, wherein the tactile sensing confidence value is determined by identifying flat surfaces and/or contrast changes exceeding a contrast change threshold.

75. The humanoid robot according to any one of claims 64 to 74 configured to store and/or access an object database comprising digital 3D models and further configured to reference tactile sensing data with at least one digital 3D model from the object database, particularly wherein the robot is configured to use said referencing during the iterative tactile sampling to determine a next surface patch to be measured by the touch sensor.

76. The humanoid robot according to any one of claims 70 to 75, being further configured

• to derive a current pose of the front face of the elastomeric component,

• to derive a shape and location of an object being touched based on the visual data, in particular using a structure from motion or simultaneous localization and mapping method,

• to adopt an orientation of the arm and/or the hand-joint arrangement based on the shape and location of the object being touched and a current pose of the touch sensor.

77. The humanoid robot according to any one of claims 70 to 76, wherein the visual sensor is a low focal length camera arranged to the hand-joint arrangement and/or on the arm in the proximity of the hand-joint arrangement.

78. The humanoid robot according to any one of claims 64 to 77, wherein the touch sensor comprises

• a touch sensor interior volume at least partly filled with the elastomeric component, wherein on the inner face the elastomeric component has a reflective coating and touching of an object with the outer face causes at least a partial reproduction of a geometric shape of a portion of the touched object on the inner face, and

• a touch sensor imaging arrangement disposed within the touch sensor interior volume and configured to provide a set of images from the objects touched by the touch sensor by imaging the inner face of the elastomeric component.

79. Humanoid robot configured for 3D surveying of an object, wherein the robot comprises a main body, two mechanically actuated legs attached to the main body at a lower portion of the main body, a mechanically actuated arm attached to the main body, and a perception sensor, wherein the legs and the arm can be swiveled independently of one another and the perception sensor is configured to generate perception data for perceiving a relative positional relationship between the object and at least part of the robot, characterized in that the robot comprises

• a large language model module, LLM-module, configured to process text input of a planned surveying task to be executed by the robot, wherein the planned surveying task involves a specific movement of the robot arm relative to the object and the LLM- module is trained to transform the text input of the planned surveying task to an output comprising robot command language,

• a robot arm instruction interpreter, configured to select robot commands out of the output of the LLM-module, • a robot arm instruction executor, configured to cause the robot to execute the robot commands selected by the robot arm instruction interpreter, resulting in an executed surveying task executed by the robot, and

• a comparison module, configured to use perception data of the perception sensor capturing a relative positional relationship between the object and at least part of the robot, particularly the arm, to compare the planned surveying task and the executed surveying task to provide a confidence value regarding achievement of the planned surveying task, wherein the robot is configured to take into account the confidence value to adapt movement of the robot when repeating the planned surveying task.

80. The humanoid robot according to claim 79, wherein the robot is configured that it comprises the teaching of one of claims 1 to 78.

81 . The humanoid robot according to claim 79 or 80, wherein the planned surveying task involves the robot to use a metrology-grade surveying sensor, wherein the metrologygrade surveying sensor is part of the robot or is held by a hand-joint arrangement of the robot arranged at a distal end of the arm remote from the main body.

82. The humanoid robot according to claim 81 , wherein the confidence value is provided based on measurement data of the metrology-grade surveying sensor.

83. The humanoid robot according to any one of claims 79 to 82, wherein the robot comprises an input functionality to provide user input to change the output of the LLM-module, wherein the robot further comprises

• a display, wherein the robot is configured to provide - on the display - feedback to an operator of the robot based on the output of the LLM-module, and/or

• a text-to-speech module configured to acoustically provide feedback to an operator of the robot based on the output of the LLM-module.

84. The humanoid robot according to any one of claims 79 to 83, comprising a speech-to-text module configured to translate voice input to the text input of the planned surveying task, particularly wherein the speech-to-text module is part of the LLM module.

85. The humanoid robot according to any one of claims 79 to 84, wherein the robot commands in robot command language comprise an instruction in a parse-able form, particularly JavaScript Object Notation or extensible markup language.

86. The humanoid robot according to any one of claims 79 to 85, wherein the comparison module is configured to

• analyze the text input of the planned surveying task to provide an input comparison parameter indicative of a movement of the robot involved by the planned surveying task,

• analyze the perception data to provide an output comparison parameter indicative of a movement of the robot involved by the executed surveying task, and

• compare the input comparison parameter and the output comparison parameter to provide the confidence value.

87. The humanoid robot according to any one of claims 79 to 86, wherein the robot is configured to take into account the confidence value

• to adapt processing of the LLM-module to provide a further output associated with the text input of the planned surveying task comprising robot commands in robot command language, wherein the robot is configured that the further output is processed by the robot arm instruction interpreter to provide a set of robot commands out of the further output, and/or

• to adapt processing of the robot arm instruction interpreter to select robot commands out of the output or the further output of the LLM-module to provide a further set of robot commands out of the output or the further output, wherein the robot is configured to carry out an adapted executed surveying task based on the robot arm instruction executor causing the robot to execute the set of robot commands and/or the further set of robot commands.

88. The humanoid robot according to any one of claims 79 to 87, wherein the comparison module is configured to

• analyze the text input of the planned surveying task to provide a planned position of the robot associated with the planned surveying task, particularly a planned position relative to the object, and

• compare the planned position and an actual position of the robot, particularly an actual position relative to the object, to provide the confidence value, particularly wherein a distance between the planned position and the actual position is determined to provide the confidence value.

89. The humanoid robot according to claim 88, wherein the robot comprises

• a locomotion instruction interpreter, configured to select robot commands out of the output of the LLM-module that provide instructions for locomotion of the robot, and a robot locomotion instruction executor, configured to cause locomotion of the robot according to the robot commands selected by the locomotion instruction interpreter, wherein the locomotion of the robot is part of the executed surveying task.

90. The humanoid robot according to claim 89, wherein the robot is configured

• to take into account the confidence value to adapt processing of the locomotion instruction interpreter to provide a selection of locomotion commands for a locomotion of the robot out of an output of the LLM-module, and

• to carry out an adapted executed surveying task based on the locomotion instruction executor causing the robot to execute the selection of locomotion commands.

91 . The humanoid robot according to any one of claims 88 to 90, wherein the comparison module is configured to determine the actual position of the robot based on at least one of

• data of an inertial measuring unit of the robot,

• image data provided by an imaging unit of the robot, and

• depth measuring data of a depth sensor of the robot, particularly wherein the perception sensor is configured to provide at least one of the image data and the depth measuring data.

92. The humanoid robot according to any one of claims 79 to 91 , configured to stop operation of the robot as a function of a defined termination criterion for successively derived confidence values.

93. The humanoid robot according to any one of claims 79 to 92, wherein the comparison module is configured to provide the confidence value by determining an actual pose of at least part of the robot and comparing the actual pose with a list of known poses of the at least part of the robot and a corresponding defined range of a permissible pose deviation for each of the known poses.

94. The humanoid robot according to any one of claims 79 to 93, wherein the comparison module is configured to provide the confidence value by determining an actual movement pattern of at least part of the robot and comparing the actual movement pattern with defined permissible movement patterns of the robot.

95. The humanoid robot according to any one of claims 79 to 94, wherein the comparison module is configured to provide the confidence value by determining a change in an environment of the robot before and after execution of the executed surveying task, particularly a position of at least part of the robot relative to the object and/or a change of positions of further objects in the environment.

96. The humanoid robot according to any one of claims 79 to 94, wherein the comparison module is configured to provide the confidence value based on an explicit feedback regarding the executed surveying task from an operator, particularly wherein at least a part of explicit feedback is provided as voice input from the operator.

97. Humanoid robot configured for 3D surveying of an object, wherein the robot comprises a main body, two mechanically actuated legs attached to the main body at a lower portion of the main body, a mechanically actuated arm attached to the main body, a hand-joint arrangement at a distal end of the arm remote from the main body, and a perception sensor, wherein the legs, the arm, and the hand-joint arrangement can be moved independently of one another and the perception sensor is configured to generate visual perception data of an environment of the robot, characterized in that the robot is configured to execute surveying tasks that involve specific movements of the robot and the use of a metrology-grade surveying sensor, wherein the metrology-grade surveying sensor is part of the robot or is held by the hand-joint arrangement, wherein the robot comprises

• a large language model module, LLM-module, configured to transform text input of a planned surveying task to an output comprising robot command language,

• a gesture recognition module configured to analyze the perception data to recognize a gesture made by an operator that is associated with the planned surveying task,

• an instruction interpreter, configured to provide robot commands based on output of the LLM-module and the gesture recognition module, and

• a robot instruction executor, configured to cause the robot to execute the robot commands, resulting in an executed surveying task executed by the robot.

98. The humanoid robot according to claim 97, wherein the robot is configured that it comprises the teaching of one of claims 1 to 96.

99. Humanoid robot configured for 3D surveying of an object, wherein the robot comprises a main body, two mechanically actuated legs attached to the main body at a lower portion of the main body, a mechanically actuated arm attached to the main body, a hand-joint arrangement at a distal end of the arm remote from the main body, and a perception sensor, wherein the legs, the arm, and the hand-joint arrangement can be moved independently of one another and the perception sensor is configured to generate visual perception data of an environment of the robot, characterized in that the robot is configured to execute surveying tasks that involve specific movements of the robot and the use of a metrology-grade surveying sensor, wherein the metrology-grade surveying sensor is part of the robot or is held by the hand-joint arrangement, wherein the robot comprises

• a gesture recognition module configured o to analyze the perception data to recognize a gesture made by an operator that is associated with a planned surveying task that involves a specific movement of the robot and the use of the metrology-grade surveying sensor, and o to translate a recognized gesture into a planned model movement sequence of a set of robot parts that are involved in the planned task,

• an instruction interpreter, configured to provide robot commands based on the planned model movement sequence,

• a robot instruction executor, configured to cause the robot to execute the robot commands, resulting in an executed surveying task executed by the robot, and

• a confirmation module, configured o to analyze perception data of the perception sensor capturing at least part of the set of robot parts to recognize an observed movement of the set of robot parts, o to translate the observed movement into an executed model movement sequence of the set of robot parts, and o to provide a confidence value regarding a matching of the planned model movement sequence and the executed model movement sequence, wherein the robot is configured to take into account the confidence value to adapt movement of the robot when repeating the planned surveying task.

100. The humanoid robot according to claim 99, wherein the robot is configured that it comprises the teaching of one of claims 1 to 98.

101. The humanoid robot according to claim 99 or 100, wherein the robot is configured to take into account the confidence value

• to adapt processing of the gesture recognition module to provide a further planned model movement sequence, wherein the robot is configured that the further model movement sequence is processed by the instruction interpreter to provide a set of robot commands, and/or • to adapt processing of the instruction interpreter to provide a further set of robot commands based on the planned model movement sequence or the further model movement sequence, wherein the robot is configured to carry out an adapted executed surveying task based on the robot instruction executor causing the robot to execute the set of robot commands and/or the further set of robot commands.

102. The humanoid robot according to any one of claims 99 to 101 , wherein the confidence value is provided based on measurement data of the metrology-grade surveying sensor and/or measurement data of the perception sensor relative to the environment.

103. The humanoid robot according to any one of claims 99 to 102, wherein the robot is configured to provide the planned model movement sequence as a scaled movement sequence or an absolute movement sequence, wherein

• for the absolute movement sequence the magnitudes of the planned model movement sequence are in one-to-one correspondence with the recognized gesture made by the operator,

• for a scaled movement sequence at least one magnitude of the planned model movement sequence is rescaled with respect to recognized gesture made by the operator by the gesture recognition module.

104. The humanoid robot according to claim 103, wherein the rescaling is carried out on the basis of the different proportions and/or joint mobility of the operator and robot.

105. Humanoid robot configured for 3D surveying of an environment, wherein the robot comprises a main body, two mechanically actuated legs attached to the main body at a lower portion of the main body, and a mechanically actuated arm attached to the main body at an arm joint at an upper portion of the main body, each of the legs comprises a track drive providing a track for track locomotion, and the robot is configured to autonomously move through the environment and to autonomously change between an upright measurement mode and a crouching measurement mode, wherein

• in the upright measurement mode, the main body is placed higher above ground than in the crouching measurement mode, contact faces of the tracks contacting the ground have smaller area sizes than contact faces of the tracks in the crouching measurement mode, and a posture of the robot is automatically balanced based on a control algorithm configured to automatically stabilize the posture based on inertial data of an inertial sensor of the robot, • in the crouching measurement mode, the robot resumes a tilt-resistant posture by having increased area sizes of the contact faces of the tracks compared to the upright measurement mode, such that the tilt-resistant posture can be maintained free of active balancing by the robot,

• the robot is configured to execute surveying tasks by a metrology-grade surveying sensor to generate 3D surveying data that provide geometric 3D information of the environment, wherein the metrology-grade surveying sensor is part of the arm or is held by the arm,

• the robot is configured to capture perception data of the environment and to carry out a classification of objects and environment areas within the perception data based on object type and area type, wherein different object types and different area types are assigned different surveying criteria for surveying with the surveying sensor, and

• the robot is configured to automatically switch between the upright measurement mode and the crouching measurement mode based on the classification.

106. The humanoid robot according to claim 105, wherein the robot is configured that it comprises the teaching of one of claims 1 to 104.

107. The humanoid robot according to claim 105 or 106, wherein the robot is configured to generate the 3D surveying data with motion correction based on compensation data of an inertial measurement unit, wherein a weighting of the compensation data for the motion correction is reduced in the crouching measurement mode compared to the upright measurement mode, particularly wherein the motion correction by the compensation data is deactivated in the crouching measurement mode.

108. The humanoid robot according to any one of claims 105 to 107, wherein the robot is configured that a range of motion of the surveying sensor relative to the robot is more restricted in the crouching measurement mode than in the upright measurement mode.

109. The humanoid robot according to any one of claims 105 to 108, wherein the robot is configured to provide at least one defined fixed holding position of the surveying sensor relative to the main body configured to provide increased mechanical stability of the surveying sensor compared to an arrangement of the surveying sensor out of the at least one fixed holding position, wherein the surveying sensor is kept in the at least one fixed holding position when generating the 3D surveying data in the crouching measurement mode, particularly the at least one fixed holding position involves a locking of the surveying sensor at the main body or a locking of the arm at the main body at a locking point of the arm that differs from the arm joint.

110. The humanoid robot according to any one of claims 105 to 109, configured that the classification takes into account 3D surveying data generated by the surveying sensor in the upright measurement mode as at least part of the perception data.

111. The humanoid robot according to any one of claims 105 to 110, comprising a visual sensor configured to provide at least part of the perception data.

112. The humanoid robot according to any one of claims 105 to 111 , configured that the crouching measurement mode provides reduced mechanical vibration than the upright measurement mode.

113. The humanoid robot according to any one of claims 105 to 112, wherein the surveying criteria define at least one of a desired measurement point density, a desired coordinate measurement accuracy, a desired measurement point pattern, a desired surveying field of view, and a desired perspective onto the object or environment area.

114. The humanoid robot according to claim 113, wherein the robot is configured to provide a command for the metrology sensor to alter a resolution and/or sampling rate and/or integration time in particular wherein the robot is configured to provide said command via a humanmachine interface of the metrology sensor.

115. The humanoid robot according to any one of claims 105 to 114, wherein the robot is configured to automatically switch an operation mode of the surveying sensor, particularly wherein the operation mode affects at least one of an emitted point pattern, an emitted point density, and a coordinate measurement accuracy.