Background
Due to the self multi-degree-of-freedom redundancy configuration characteristics of the multi-legged robot, when the multi-legged robot is applied to a non-structural environment, the multi-legged robot has stronger environmental adaptability and task operability compared with other robots by means of a walking mode that swing legs and supporting legs are switched successively and high stability margin under multi-legged support. However, in a complex and changeable working scene, as the multi-legged robot may encounter practical situations such as local terrain sudden change, instability action recovery, target object interaction and the like in the walking process, a targeted teleoperation strategy needs to be formulated at this time, so as to ensure that the teleoperation system of the multi-legged robot has good controllability and environmental adaptability. Therefore, the multi-target multi-dimensional teleoperation system of the multi-legged robot can be designed by taking the working condition of the obstacle environment as the background.
Aiming at the problems, the teleoperation scheme which can regulate and control the single-leg operation and the body translation of the multi-legged robot in a coordinated manner is designed with the aim of maximally improving the operational capability of the multi-legged robot on the premise of ensuring the stability margin of the body. The method mainly solves the problem that the multi-degree-of-freedom coupling effect in the translation process of the machine body has influence on the stable tracking of the system and the influence of unknown model parameters, variable contact states and unmeasurable operating force on the tracking precision and the force transparency of the system during single-leg operation. The designed multi-legged robot teleoperation system adopts a teleoperation mode of multiple masters and single slaves, and the whole teleoperation system is decomposed into two subsystems: the machine layer and the single-leg layer respectively provide two sets of control frameworks and algorithms: the machine layer adopts a control framework based on position tracking, and the single-leg layer adopts a 4-channel control framework which integrates a nonlinear force estimation algorithm and a self-adaptive robust control strategy. In addition, control law parameters of a teleoperation subsystem of the machine body layer are reasonably designed based on the absolute stability criterion of the multi-degree-of-freedom coupling system, and the adaptive law of the parameters of the single-leg layer controller is determined by applying the Lyapunov function.
Disclosure of Invention
The invention aims to solve the problems of insufficient controllability and environmental adaptability of a conventional multi-foot robot teleoperation system when the multi-foot robot is in an obstacle environment (collapse, obstacle and target object exist in a local environment), and provides a teleoperation system and a control method capable of cooperatively regulating and controlling single-leg operation and body translation of the multi-foot robot in order to give full play to the motion potential of the multi-foot robot and overcome the influence of complicated and changeable environmental constraints on the working capacity of the multi-foot robot.
The teleoperation system and the control method for cooperatively regulating and controlling single-leg operation and body translation of the multi-legged robot are mainly divided into four parts: making a teleoperation scheme of the multi-legged robot in an obstacle environment; establishing a remote operation system of the multi-legged robot in an obstacle environment; modeling, controlling architecture and controlling algorithm of the multi-legged robot body-level teleoperation subsystem; modeling, control architecture and control algorithm design of the single-leg layer teleoperation subsystem of the multi-legged robot.
The teleoperation system and the control method capable of cooperatively regulating and controlling single-leg operation and body translation of the multi-legged robot specifically comprise the following steps:
step 1: according to the structural characteristics of multi-degree-of-freedom redundancy of the multi-legged robot and the restriction of non-structural contact conditions in an obstacle environment, a teleoperation scheme based on cooperative regulation and control of body translation and single-leg operation is designed for improving the controllability and adaptability of a teleoperation system of the multi-legged robot;
step 2: according to the teleoperation scheme formulated in the step 1, a slave end system in the interaction process of the multi-legged robot and the environment is regarded as an operable mechanical arm with a mobile base, so that components of the teleoperation system of the multi-legged robot in the obstacle environment are constructed, and the teleoperation scheme specifically comprises the following steps: a body layer teleoperation subsystem and a single-leg layer teleoperation subsystem;
and step 3: determining a control architecture of the body layer teleoperation subsystem according to the teleoperation system constructed in the step 2, establishing a model of a master end and a slave end of the body layer teleoperation subsystem, designing a control algorithm of the body layer teleoperation subsystem on the basis, and solving control law parameters through an absolute stability criterion of the multi-degree-of-freedom coupling system;
and 4, step 4: and 2, determining a control architecture of the single-leg-layer teleoperation subsystem according to the teleoperation system constructed in the step 2, establishing a model of a master end and a slave end of the single-leg-layer teleoperation subsystem, designing a control algorithm of the single-leg-layer teleoperation subsystem on the basis, and designing a control law by utilizing a Lyapunov function.
The invention has the following beneficial effects:
on one hand, by designing a teleoperation scheme for the cooperative regulation of the single-leg operation and the body translation, the body translation amount of the slave-end robot and the position instruction of the body layer main-end robot are tracked and matched, and meanwhile, the displacement amount of the tail end of the operable leg of the slave-end robot and the position instruction of the single-leg layer main-end robot are tracked and mapped, so that the working space and the working capacity of the multi-leg robot are improved to the maximum extent while the body stability margin is maintained; on the other hand, a bilateral control framework of the body layer teleoperation subsystem of the multi-legged robot is designed based on a position tracking mode, stable tracking of a machine body with a multi-freedom-degree coupling effect is guaranteed, meanwhile, a bilateral control mode of the single-leg layer teleoperation subsystem of the multi-legged robot is designed based on a force-position mixed control mode, high position tracking precision of the system is guaranteed, and good force transparency is achieved.
Detailed Description
Embodiment mode 1: in the teleoperation system capable of cooperatively regulating and controlling single-leg operation and body translation of the multi-legged robot according to the embodiment, firstly, a teleoperation scheme of the system needs to be determined, and the teleoperation scheme of the multi-legged robot in an obstacle environment includes: designing an overall control flow, a mapping mode and a force feedback mode of the teleoperation system, wherein the overall control flow of the teleoperation system is designed based on different working scenes and task requirements; the teleoperation mode of the system adopts a multi-master-single-slave control mode, wherein one master-end robot is responsible for controlling the body translation amount of the slave-end robot, and the other master-end robot is responsible for controlling the displacement amount of the tail end of the operable leg of the slave-end robot.
Embodiment mode 2: the teleoperation system capable of cooperatively regulating and controlling single-leg operation and body translation of the multi-legged robot in the embodiment divides the teleoperation system of the multi-legged robot into two parts: a body layer teleoperation subsystem and a single-leg layer teleoperation subsystem; referring to the attached fig. 1, the composition of the teleoperation system of the multi-legged robot specifically includes an operation end 1 and an environment end 2, a body layer master end robot 3, a body layer master end controller 4, a body layer slave end controller 5, a body 6 of the slave end robot, a single-leg layer master end robot 7, a single-leg layer master end controller 8, a single-leg layer slave end controller 9, an operable leg 10 of the slave end robot, and a communication port 11.
Embodiment mode 3: in this embodiment, the overall control flow of the teleoperation system capable of cooperatively regulating and controlling the single-leg operation and the body translation of the multi-legged robot is further described with reference to embodiment 1, and in this embodiment, the proposed control flow can be described according to three working scenarios with reference to the overall control flow of the teleoperation system of the multi-legged robot shown in fig. 2: scene 14: when a local surmountable obstacle appears in the working space of the slave-end robot 12, the operator can manually intervene on the operable leg of the slave-end robot through the single-leg layer master-end robot and control the slave-end robot to interact with the target obstacle 13 according to the intention of the operator; scene 15: after the operable leg of the slave-end robot is stepped on the surface of the target obstacle 13, since the leg is far away from the target foot-drop point which is planned autonomously, the slave-end robot 12 needs to maintain the stability margin of the body in the original movement direction, so that the operator can control the body translation amount of the slave-end robot by controlling the body layer master-end robot. Scene 16: at this time, the body and operable legs of the slave-end robot meet the crossing condition of the target obstacle 13, and the operator can regulate and control the single-leg-layer master-end robot again, so that the single-leg crossing of the local obstacle is completed, and a new foot-falling position is found.
Embodiment 4: in this embodiment, referring to the structure of the robot at the main end of the robot body layer shown in fig. 3, a handle 17 of the robot at the main end of the robot body layer has 3 degrees of freedom in a working space thereof, which are a position command 18 along the X-axis direction in the working space of the handle, a position command 19 along the Y-axis direction in the working space of the handle, and a position command 20 along the Z-axis direction in the working space of the handle, respectively;
referring to the structural model of the slave end robot body shown in fig. 4, wherein the slave end robot is electrically driven by a hexapod robot, the hexapod robot consists of a body and six legs with the same configuration, when the slave end robot is in a working state of body translation, all 18 joints including 15 supporting joints 21 of supporting legs and 3 operating joints 22 of operable legs are responsible for jointly pushing the body of the slave end robot to complete translation movement, and the controlled degrees of freedom of the body of the slave end robot comprise: the translation amount 23 along the X-axis direction of the body coordinate system, the translation amount 24 along the Y-axis direction of the body coordinate system, and the translation amount 25 along the Z-axis direction of the body coordinate system;
the teleoperation mode of the teleoperation subsystem of the designed machine body layer is as follows: a position instruction 18 of the robot at the main end of the machine body layer along the X-axis direction in the handle working space maps the translation 23 of the machine body of the robot at the slave end along the X-axis direction of the machine body coordinate system; a position instruction 19 of the robot at the main end of the machine body layer along the Y-axis direction in the handle working space maps the translation amount 24 of the machine body of the robot at the auxiliary end along the Y-axis of the machine body coordinate system; the position command 20 of the robot body main end robot handle along the Z-axis direction in the handle working space maps the translation 25 of the robot body of the slave end robot along the Z-axis of the robot body coordinate system.
Embodiment 5: in this embodiment, reference is made to the structure of a single-leg layer main end robot shown in fig. 5, wherein the single-leg layer main end robot has 3 degrees of freedom, and the tail end 26 can move in the working space thereof along the X-axis direction to output a position command 27, move in the Y-axis direction to output a position command 28, and move in the Z-axis direction to output a position command 29;
referring to the single-leg operation diagram of the slave end robot shown in fig. 6, when the slave end robot is in the working state of single-leg operation, the controlled degrees of freedom of the operable leg tip 30 include: a displacement 31 in the X-axis direction of the end coordinate system, a displacement 32 in the Y-axis direction of the end coordinate system, and a displacement 33 in the Z-axis direction of the end coordinate system;
the teleoperation mode of the designed single-leg layer teleoperation subsystem is as follows: the position command 27 output by the single-leg layer main end robot tail end 26 in the moving direction of the X-axis in the working space thereof maps the displacement 31 of the operable leg tail end 30 of the slave end robot in the X-axis direction of the tail end coordinate system; the position command 28 output by the single-leg layer main end robot end 26 moving in the Y-axis direction within its working space maps the displacement 32 of the slave end robot operable leg end 30 in the Y-axis direction of the end coordinate system; the position command 29 output by the one-leg layer main end robot tip 26 moving in the Z-axis direction within its workspace maps the amount of displacement 33 of the slave end robot operable leg tip 30 in the Z-axis direction of the tip coordinate system.
Embodiment 6: in this embodiment, assuming that there is no relative slip in the contact between the operable leg end and the target object, the force feedback mode is designed to be expressed as: when the translation amount of the slave robot body is regulated, the position tracking error in the translation process of the body is simulated in the body layer main end controller, and the position tracking error is fed back to the corresponding touch guidance force of an operator. When the displacement of the tail end of the operable leg of the slave-end robot is regulated, the operable leg is in a motion state in which the free space and the constraint space are mutually switched, so that the position tracking error of the leg in the free motion process and the environmental force acting on the tail end in the contact process need to be sent to the single-leg-layer master-end controller in real time, and an operator can really sense the working state of the operable leg of the slave-end robot in such a way.
Embodiment 7: the teleoperation control method capable of cooperatively regulating and controlling single-leg operation and body translation of the multi-legged robot specifically comprises a control framework and a control method for designing a body layer teleoperation subsystem, wherein an operator quantizes a position control instruction of a body of the slave-end robot through a body layer master-end robot, and the body layer slave-end controller sends body motion information to the body layer master-end controller to form a touch guiding force and feeds the touch guiding force back to the operator; designing a control architecture and a control method of a single-leg-layer teleoperation subsystem, quantizing a position control instruction of the tail end of an operable leg of a slave-end robot by an operator through a single-leg-layer master-end robot, and sending position information and force information of the operable leg to the single-leg-layer master-end controller by the single-leg-layer slave-end controller to form a touch guiding force and feeding the touch guiding force back to the operator; the bottom controller of the remote operation system of the multi-legged robot is designed to realize the position control instruction of the automatic tracking body of the angular speed and the position control instruction of the operable leg tail end which are required to be output from the electric drive joint of the end robot.
Embodiment mode 8: the present embodiment is further described with respect to a bottom controller in a teleoperation control method capable of cooperatively regulating and controlling single-leg operation and body translation of a multi-legged robot in the specific embodiment 7, in the present embodiment, a joint control algorithm is designed in the bottom controller;
the bottom layer controller is embedded with a single-leg kinematic model of the slave end robot, target rotation angles of all supporting leg joints solved by the slave end controller of the robot layer are obtained according to target positions of tail ends of operable legs solved by the slave end controller of the robot layer, target rotation angles required to be output by 3 joints of the operable legs are obtained through inverse Jacobian matrix operation, the joint rotation angles of all the solved legs are integrated, and motors built in the leg joints are actually driven to complete work tasks appointed by the slave end robot through a designed closed-loop PID feedback control algorithm of the joint layer.
Embodiment mode 9: in the present embodiment, referring to fig. 7, an operator controls a robot at a main end of a robot body layer to output a position command in three directions, the position command is converted by an impedance control model 35 of the robot at the main end of the robot body layer, and then, in combination with a control law 36 at a main end of the robot body layer, the position command sent by the robot at the main end of the robot body layer is converted into a translation command expected by the robot at a slave end and sent to a body of the robot at the slave end, an expected translation quantity is converted into an actual position of the body of the robot at the slave end by a kinematic model 37 of the body-single leg coupling action of the robot at the slave end, and under the action of a control law parameter 38 at the slave end of the body layer, the expected translation quantity passes through an impedance model 39 at an environment end of the body layer, the external acting force is converted into an external acting force which causes a position tracking difference to appear on the slave robot body, the external acting force is simulated in the body layer main end controller and converted into tactile force information, and the updated body expected command is iterated in the body layer main end robot 3 through an impedance model of the body layer of the arm of the operator.
Embodiment 10: the present embodiment is further described with respect to a method for controlling a body layer teleoperation subsystem in a teleoperation control method capable of cooperatively regulating and controlling single-leg operation and body translation of a multi-legged robot in embodiment 7, where the method specifically includes the following steps:
step 1: performing dynamic modeling on the robot at the main end of the body layer, so as to establish an impedance control model of the robot at the main end of the body layer;
step 2: establishing a slave-end robot body kinematic model considering the coupling effect by analyzing the coupling effect existing in the process of cooperatively regulating and controlling single-leg operation and body translation of the slave-end robot in the obstacle environment;
and step 3: designing a control algorithm of the teleoperation subsystem of the body layer:
step 3.1: establishing an impedance control model of a teleoperation subsystem of a body layer based on a dual-port network technology in the circuity, mapping a position instruction of a master robot of the body layer with an expected translation amount of a body of a slave robot through a designed control rule parameter, determining motion characteristics of the body of the slave robot along three directions (transverse, longitudinal and vertical) of a coordinate axis of the body under the action of single-leg coupling through a kinematic model of the body of the slave robot planned in the step 2, converting the expected translation amount of the body into expected rotation angles of all joints including all supporting legs and operable legs, supporting the body to translate through a motor in each joint, and obtaining the actual position of the body through an impedance model 39 of the environment end of the body layer;
step 3.2: the actual translation amount of the robot body is differed from the expected translation amount in the slave-end controller of the robot body layer, then the external acting force which causes the difference in the position tracking process is simulated in the master-end controller of the robot body layer to form tactile force information, a force feedback motor in the master-end robot of the robot body layer is controlled to output the tactile force, an operator senses the force information through the master-end robot of the robot body layer, and the information is converted into a new position instruction through the impedance characteristic of the operator and is output to the master-end robot of the robot body layer again;
and 4, step 4: the coupling influence between the terminals (an operation end and an environment end) of the multi-degree-of-freedom body layer teleoperation subsystem and a master-slave control law is analyzed, the control law parameters of the body layer teleoperation subsystem are solved through the absolute stability criterion of the multi-degree-of-freedom coupling system, and the stability of the designed body layer teleoperation subsystem is ensured.
Embodiment mode 11: the present embodiment is further described with reference to the control architecture of the single-leg layer teleoperation subsystem in the teleoperation control method capable of cooperatively regulating and controlling single-leg operation and body translation of the multi-legged robot described in embodiment 7, in the present embodiment, referring to fig. 8, an operator outputs a position command in three directions by controlling the end of the single-leg layer master-end robot, estimates the operation force applied to the single-leg layer master-end robot by the operator based on a non-linear force observer 44 designed in the single-leg layer master-end controller, under the action of an operation force gain 41 of the single-leg layer master-end system, combines with the single-leg layer master-end system local control law 41, converts the position command through an impedance control model 45 of the single-leg layer master-end robot, thereby obtaining a desired displacement amount of the end of an operable leg and sending the displacement amount to the single-leg layer slave-end controller, and uses an adaptive robust controller 48 designed in the single-leg layer slave-end controller, eliminating the influence of parameter unknown and external unknown disturbance in the coupling dynamic model of the operable leg of the slave-end robot, on the basis, forming force-position hybrid control in the single-leg layer slave-end controller by combining the operating force estimated by the nonlinear force estimator in the single-leg layer master-end controller with the operating force gain 46 of the single-leg layer slave-end system and the local control law 47 of the single-leg layer slave-end system, and converting the expected displacement into the actual position of the tail end of the operable leg of the slave-end robot through the coupling dynamic model 49 of the operable leg of the slave-end robot and the impedance model 50 of the environment end of the single-leg layer; the method comprises the steps that the operable legs of a slave robot are mutually switched under two states of free motion and contact constraint to cause position tracking and force tracking of the tail end to fluctuate, force tracking information is sent to a single-leg-layer main-end controller under the action of a single-leg-layer teleoperation subsystem control law 43 and is directly used for force closed-loop control, then the position tracking information and the force tracking information of the master end and the slave end of the single-leg-layer teleoperation subsystem are fused by combining a control law 42 of the single-leg-layer teleoperation subsystem, all information used for closed-loop force control is centrally processed in the single-leg-layer main-end controller, the processed force feedback information is simulated and corresponding tactile force information is output, and an updated expected command of the tail end of the operable leg is iterated in a single-leg-layer main-end robot 7 through an impedance model 50 of an arm of an operator on a single leg layer.
Embodiment mode 12: the present embodiment is further described with respect to a control method of a single-leg layer teleoperation subsystem in a teleoperation control method capable of cooperatively regulating and controlling single-leg operation and body translation of a multi-legged robot in specific embodiment 7, where the control method specifically includes the following steps:
step 1: performing dynamic modeling on the single-leg layer main-end robot, so as to establish an impedance control model of the single-leg layer main-end robot;
step 2: establishing a dynamic model of the operable legs of the slave-end robot under the coupling effect by analyzing the coupling effect existing in the process of cooperatively regulating and controlling the single-leg operation and the body translation of the slave-end robot in the obstacle environment;
and step 3: designing a control algorithm of the single-leg layer teleoperation subsystem:
step 3.1: establishing an impedance control model of a single-leg layer teleoperation subsystem based on a dual-port network technology in the circuity, mapping a position instruction of a single-leg layer master-end robot with a desired position of an operable leg tail end of a slave-end robot through designed control law parameters, determining motion characteristics of the operable leg tail end of the slave-end robot along three directions (transverse direction, longitudinal direction and vertical direction) of a working space of the slave-end robot when coupling action of body translation and contact environment exists, converting the desired displacement amount of the operable leg tail end into a desired rotation angle of 3 joints of the operable leg through a coupling dynamic model of the operable leg of the slave-end robot planned in step 2, an operating force estimation algorithm in a single-leg layer master-end controller and an adaptive robust control algorithm in the single-leg layer slave-end controller, and driving the operable leg to swing and contact through a motor built in the joint, then obtaining the actual position of the tail end of the operable leg through an environment end impedance model of the single-leg layer;
step 3.2: fusing a position tracking difference value of the tail end of an operable leg and a contact force of an environment end acting on the tail end in a single-leg layer slave-end controller, simulating the fused position tracking information and force tracking information in a single-leg layer master-end controller to form tactile force information, controlling a force feedback motor in a single-leg layer master-end robot to output the tactile force, sensing the force information by the single-leg layer master-end robot by an operator, converting the information into a new position instruction through the impedance characteristic of the operator, and outputting the new position instruction to the single-leg layer master-end robot again;
and 4, step 4: control law parameters of the single-leg layer teleoperation subsystem are designed based on the attributes of a force-position hybrid control 4-channel teleoperation architecture, adaptive laws of parameters to be estimated in the single-leg layer teleoperation subsystem are constructed through a Lyapunov function, and the stability and the transparency of the designed single-leg layer teleoperation subsystem are ensured.