JP2006000966A - Robot device and its control method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a control method for a robot device improving entertaining nature and usability. <P>SOLUTION: In this robot device, the rotation of a motor serving as a power source for driving a movable part is controlled according to action contents to be carried out, and the rotating state of the motor is detected. Based on the detected rotating state of the motor, the magnitude of external force applied to the movable part is estimated. Based on the estimated result of the magnitude of external force, the rotation of the motor is controlled so that the magnitude of external force and the displacement quantity in the rotating state of the motor due to the external force are in the proportional relation with the preset adjustment gain as a proportional constant, and the adjustment gain set to a control means is changed according to the situation. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a robot apparatus and a control method therefor, and is suitably applied to, for example, a humanoid type robot apparatus.

  In recent years, research and development of humanoid robots that imitate human forms have been carried out in many companies and various research institutions.

  In this case, in this type of robot apparatus, the structure of the “hand” is very important in performing various operations, and when a movement closer to a human is desired, a complicated operation such as grasping, picking or grasping an object is performed. It is required to have a structure that can reliably perform the movement, and various devices are also required for the movement.

Therefore, conventionally, as a hand structure of such a humanoid type robot and a method for controlling the hand of the robot, the gripping ability is improved by making the finger a three-layer structure of a skeleton layer, a flexible layer and a surface layer having different hardnesses. It has been proposed by the applicant of the present patent application that the object can be grasped reliably and appropriately by performing compliance control on the movement of the finger and such a finger (see, for example, Non-Patent Document 1).
Japanese Patent Application No. 2002-341047

  By the way, in the conventional robot disclosed in Non-Patent Document 1 described above, the softness of the finger portion with respect to the external force by the compliance control is fixed, and therefore the softness of the finger portion with respect to the external force is set to be soft. Can not perform actions and work that require high rigidity to the finger, such as pressing a switch, and conversely has a complicated outer shape when the softness of the finger against external force is set firmly There is a problem that it is impossible to perform an action or work that requires low rigidity of the finger part, such as holding an external object softly following the finger part on the surface.

  However, in consideration of future use environments of robots, it is practically sufficient to handle both actions and tasks that require high rigidity for such fingers and actions and tasks that require low rigidity. Appearance of a robot to be obtained is required, and conversely, if such a robot can be constructed, it is considered that the usefulness of the robot can be improved by expanding the range of uses of the robot.

  On the other hand, in conventional robots, no consideration is given to finger protection that plays an important role in performing various operations as described above. In general, robot fingers are slender and structurally broken. It easily breaks due to a collision with the floor when the robot falls or a collision with an external object when you move your hand vigorously. .

  The present invention has been made in consideration of the above points, and intends to propose a robot apparatus and a control method thereof that can improve usability.

  In order to solve such a problem, in the present invention, in the robot apparatus, in accordance with a command given from the host controller, a control means for controlling the rotation of the motor as a power source for driving the movable part, and the rotation state of the motor are detected. And an external force estimating means for estimating the magnitude of the external force acting on the movable part based on the detection result of the sensor means, and the control means determines the magnitude of the external force based on the estimation result of the external force estimating means. The motor rotation is controlled so that the amount of displacement in the rotation state of the motor due to the external force has a proportional relationship with a preset adjustment gain as a proportional constant, and the host controller is set in the control means according to the situation. The adjustment gain was changed.

  As a result, in this robot apparatus, the softness of the movable part with respect to an external force by so-called compliance control can be switched according to the situation.

  According to the present invention, in the control method of the robot apparatus, the rotation of the motor as the power source for driving the movable part is controlled according to the action content to be executed, and the first rotation state of the motor is detected. A step, a second step of estimating the magnitude of the external force acting on the movable part based on the detected rotational state of the motor, and a magnitude of the external force based on the estimation result of the magnitude of the external force, and the external force The third step of controlling the rotation of the motor is set in the control means according to the situation so that the displacement amount in the rotation state of the motor has a proportional relationship with a preset adjustment gain as a proportional constant. And a fourth step for changing the adjustment gain.

  As a result, according to the control method of the robot apparatus, the softness of the movable part with respect to an external force by so-called compliance control can be switched according to the situation.

  According to the present invention, in a robot apparatus having a movable part that can be displaced, a control means for controlling the rotation of a motor as a power source for driving the movable part in response to a command given from the host controller, and the rotational state of the motor Sensor means for detecting the external force, and external force estimation means for estimating the magnitude of the external force acting on the movable part based on the detection result of the sensor means, and the control means determines the external force based on the estimation result of the external force estimation means. The rotation of the motor is controlled so that the magnitude and the amount of displacement of the rotation state of the motor due to the external force have a proportional relationship with a preset adjustment gain as a proportional constant, and the host controller controls the control means according to the situation. By changing the set adjustment gain, the softness of the movable part against external force by so-called compliance control is switched according to the situation. Bets can be, thus possible to realize a robot apparatus capable of improving the usability.

  According to the invention, in the control method of the robot apparatus having the movable part that can be displaced, the rotation of the motor as the power source for driving the movable part is controlled according to the content of the action to be performed, Based on the first step of detecting the rotation state, the second step of estimating the magnitude of the external force acting on the movable part based on the detected rotation state of the motor, and the estimation result of the magnitude of the external force, A third step for controlling the rotation of the motor so that the magnitude of the external force and the amount of displacement in the rotational state of the motor by the external force have a proportional relationship with a preset adjustment gain as a proportional constant; The fourth step of changing the adjustment gain set in the control means is provided, so that the softness of the movable part against the external force by so-called compliance control is provided. Depending can be switched in, thus possible to realize a control method for a robot apparatus capable of improving the usability.

  Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

(1) Configuration of Robot According to this Embodiment In FIGS. 1 and 2, 1 indicates a bipedal walking robot according to this embodiment as a whole, and a head unit 3 is disposed above the body unit 2. In addition, arm units 4A and 4B having the same configuration are respectively disposed on the upper left and right of the body unit 2, and leg units 5A and 5B having the same structure are disposed on the lower left and right of the body unit 2, respectively. It is configured by being attached at a predetermined position.

In the torso unit 2, a frame 10 that forms the upper part of the trunk and a waist base 11 that forms the lower part of the trunk are connected via a hip joint mechanism 12. By driving the actuators A ₁ and A ₂ of the fixed hip joint mechanism 12, the upper part of the trunk can be independently rotated around the orthogonal roll axis 13 and the pitch axis 14 shown in FIG. Has been made.

The head unit 3 is attached to the center of the upper surface of the shoulder base 15 fixed to the upper end of the frame 10 via a neck joint mechanism 16, and the actuators A ₃ and A ₄ of the neck joint mechanism 16 are respectively connected to the head unit 3. By being driven, it can be rotated independently around the orthogonal pitch axis 17 and yaw axis 18 shown in FIG.

Furthermore arm units 4A, 4B are attached to the left and right shoulder base 15 respectively via a shoulder joint mechanism 19, FIG by driving corresponding to each actuator A _5, A ₆ of the shoulder joint mechanism 19 respectively 3 can be rotated independently around the orthogonal pitch axis 20 and roll axis 21 shown in FIG.

In this case, each of the arm units 4A, 4B, the actuator A ₈ to form a forearm via an elbow joint mechanism 22 is connected to each output shaft of the actuator A ₇ forming the upper arm, the tip end of the forearm It is configured by attaching the hand portion 23.

The arm units 4A, in 4B, the forearm is rotated around the yaw axis 24 shown in FIG. 3 by driving the actuator A _7, pitch indicating the forearm in Fig. 3 by driving the actuator A ₈ Each of them can be rotated around an axis 25.

On the other hand, each leg unit 5A, 5B is attached to the waist base 11 at the lower part of the trunk via the hip joint mechanism 26, and the actuators of the corresponding hip joint mechanism 26 are respectively A _{9 to} A _11. By driving each of them, the yaw axis 27, the roll axis 28 and the pitch axis 29 which are orthogonal to each other shown in FIG.

  In this case, each leg unit 5A, 5B is connected to a lower end of a frame 30 that forms a thigh, a frame 32 that forms a lower leg through a knee joint mechanism 31, and to the lower end of the frame 32. The foot portion 34 is connected via an ankle joint mechanism 33.

Thus leg units 5A, in 5B, by driving the actuator A ₁₂ which forms a knee joint mechanism 31, it is possible to rotate the lower leg around the pitch axis 35 shown in FIG. 3, also ankles By driving the actuators A ₁₃ and A ₁₄ of the joint mechanism 33, the foot 34 can be rotated independently around the orthogonal pitch axis 36 and roll axis 37 shown in FIG.

  On the other hand, on the back side of the waist base 11 that forms the lower part of the trunk of the trunk unit 2, as shown in FIG. 4, a main control unit 40 that controls the operation of the entire robot 1, a power supply circuit, a communication circuit, and the like. A control unit 42 in which the peripheral circuit 41 and the battery 45 (FIG. 5) are housed in a box is provided.

  The control unit 42 includes sub-control units 43A to 43A disposed in the constituent units (the body unit 2, the head unit 3, the arm units 4A and 4B, and the leg units 5A and 5B). It is connected to 43D and can supply a necessary power supply voltage to these sub-control units 43A to 43D and can communicate with these sub-control units 43A to 43D.

The sub-control units 43A to 43D are connected to the actuators A _{1 to} A _{14 in} the corresponding constituent units, respectively, and the actuators A _{1 to} A ₁₄ in the constituent units are given from the main control unit 40. It can be driven to a designated state based on various control commands.

  Further, as shown in FIG. 5, the head unit 3 includes an external device including a CCD (Charge Coupled Device) camera 50 that functions as an “eye” of the robot 1, a microphone 51 that functions as an “ear”, a touch sensor 52, and the like. A sensor unit 53 and a speaker 54 functioning as a “mouth” are disposed at predetermined positions, and an internal sensor unit 57 including a battery sensor 55 and an acceleration sensor 56 is disposed in the control unit 42. Yes.

  Then, the CCD camera 50 of the external sensor unit 53 captures the surrounding situation and sends the obtained image signal S1A to the main control unit, while the microphone 51 provides “walk” and “down” given as voice input from the user. ”Or“ Follow the ball ”is collected, and the audio signal S1B thus obtained is sent to the main control unit 40.

  1 and 2, the touch sensor 52 is provided in the upper part of the head unit 3, and detects the pressure received by the physical action such as “blow” or “slap” from the user. The detection result is sent to the main control unit 40 as a pressure detection signal S1C.

  Further, the battery sensor 55 of the internal sensor unit 57 detects the remaining energy of the battery 45 at a predetermined period, and sends the detection result to the main control unit 40 as a remaining battery level detection signal S2A, while the acceleration sensor 56 has three axes. The acceleration in the direction (x-axis, y-axis, and z-axis) is detected at a predetermined cycle, and the detection result is sent to the main control unit 40 as an acceleration detection signal S2B.

  The main control unit 40 includes an image signal S1A, an audio signal S1B, a pressure detection signal S1C, and the like (hereinafter collectively referred to as an external sensor) supplied from the CCD camera 50, microphone 51, touch sensor 52, and the like of the external sensor unit 53, respectively. (Referred to as the signal S1), the remaining battery level detection signal S2A and the acceleration detection signal S2B supplied from the battery sensor 55 and the acceleration sensor of the internal sensor unit 57, respectively (hereinafter collectively referred to as the internal sensor signal S2). Based on the above, the situation around and inside the robot 1, the instruction from the user, the presence / absence of the action from the user, and the like are determined.

The main control unit 40 determines the action to be continued based on the determination result, the control program stored in the internal memory 40A in advance, and various control parameters stored in the external memory 58 loaded at that time. Then, the control command based on the determination result is sent to the corresponding sub-control units 43A to 43D. As a result, based on this control command, the corresponding actuators A _{1 to} A ₁₄ are driven under the control of the sub-control units 43A to 43D, thus swinging the head unit 3 up and down, left and right, Actions such as raising the unit units 4A and 4B or walking are expressed by the robot 1.

  At this time, the main control unit 40 outputs a sound based on the sound signal S3 to the outside by giving a predetermined sound signal S3 to the speaker 54 as necessary, or functions as an “eye” in appearance. By outputting a drive signal to the LED provided at a predetermined position of the unit 3, this is blinked.

  In this way, the robot 1 can behave autonomously based on the surrounding and internal conditions, instructions from the user, presence / absence of actions, and the like.

(2) Configuration of Hand 23 in Robot 1 (2-1) Schematic Configuration of Hand 23 Next, the configuration of the hand 23 in the robot 1 will be described.

As shown in FIGS. 6 (A) and 6 (B), the robot 23 has a hand portion 23 having first to fifth finger portions 60 ₁ to 60 ₁ corresponding to a human thumb, index finger, middle finger, ring finger, and little finger, respectively. 60 ₅ is provided.

In this case, the first finger portion 60 ₁ is constituted by a fingertip portion 61 ₁ and the Yubimoto portion 61 ₂ is refracted rotatably connected toward the inside of the first knuckle 61 ₃ via the hand portion 23 are, the lower end portion of the Yubimoto portion 61 _2, the inner surface of the hand main body 62 (hereinafter referred to as the plane of the hand) lower near the right end position in 62A (right hand) or lower near the left end position (left hand), It is bent rotatably connected toward the inside of the hand portion 23 through the second knuckle 61 _4.

The second to fifth fingers ₆₀ 2-60 ₅ is refracted rotatably connected toward the fingertip portion 62 ₁ and the middle finger portion 62 _2, respectively to the inside of the hand portion 23 through the first knuckle 62 ₃ Rutotomoni, is constituted by a finger central 62 ₂ and Yubimoto 62 ₄ is refracted rotatably connected toward the inside of the second knuckle 62 ₅ through the hand unit 23, the lower end of Yubimoto 62 ₄ parts have been bent rotatably connected toward the inside of the hand portion 23 through the third finger joint 62 ₆ to the tip of the hand body portion 62.

Thereby, in this robot 1, by extending or bending the _first to fifth finger portions 60 _{1 to} 60 ₅ , the hand portion is opened as shown in FIG. 6, or the hand portion as shown in FIGS. 23 can be closed, and thus, for example, the objects 64 and 65 such as balls can be grasped as shown in FIGS. 9 and 10, and the object 66 such as paper and thin plates can be picked as shown in FIG. Has been made.

In this case the second to fifth fingers ₆₀ 1 to 60 _5, as shown in FIG. 6, in opened hand, these second to fifth fingers ₆₀ 1 to 60 ₅ is the hand body 62 Is attached to the hand main body 62 so as to extend in the radial direction at a constant angular interval (for example, 15 °). The first finger portion 60 _1, as shown in FIG. 7, attached to the hand body portion 62 so as to intersect with a certain angle theta _th its center line K ₁ is the center line K ₂ of the third finger ing.

As a result, in the robot 1, the first to fifth finger portions 60 _{1 to} 60 ₅ can hold the objects 64 and 65 so as to be wrapped from a plurality of directions as shown in FIGS. 9 and 10, for example. Thus, it is possible to reliably hold a wide range of sizes of objects from large objects to small objects.

Further, according the first angle theta _th fingers 60 _1, when formed by bending the first to fifth fingers ₆₀ 1 to 60 ₅ of the state grasping the object as shown in FIG. 7, the first tip of the finger portion 60 ₁ are selected in the angle of contact respectively with the second tip portion of the finger portion 60 ₂ and the third finger portion 60 ₃ of the tip, thus when picking an object, Figure 11 as reliably at least a first finger portion 60 ₁ of the front end portion, a second finger portion 60 ₂ of the tip, the object 67 by the three points and the third finger portion 60 ₃ of the tip of the To be able to hold on.

In the case of this embodiment, the lengths of the _first to fifth finger portions 60 _{1 to} 60 ₅ are selected to be different from each other, whereby the first to fifth finger portions 60 _{1 to} 60 60 are selected. _An object can be gripped while contacting ₅ in a wide range.

(2-2) Specific Configuration of Hand Part 23 Here, in practice, the hand main body part 62 is configured such that the flat surface 62A of the hand has a substantially round rectangular shape as a whole, as shown in FIG. a lower portion including the first mounting position of the finger unit 60 _1, the hill portion _62A 1 of the tip portion and a convex shape which accounts for approximately 3 area 4 of the total plane 62A of the hand, 62A ₃ (FIG. 12 ( in a) with the upper and lower portions than the hatched portion) is provided, in these tip of lands 62A ₁ and lower lands 62A ₃ between concave recess 62A ₂ (Figure 12 (a) (Hatched portion) is provided.

In this case, as it is apparent from FIG. 12, whereas between lands 62A ₁ and the recessed portion 62A ₂ of the tip are connected by a smooth curved surface, the bottom between the hill portion 62A ₃ and the recessed portion 62A ₂ Is provided with a step 62B, whereby the object can be stably placed on the hand plane 62A, and the object can be stably grasped.

  Further, a surface material 63 made of a material having a large coefficient of friction such as a rubber material is attached to the hand flat surface 62A of the hand main body 62, whereby the surface material 63, the object, The object can be held in the plane 62A of the hand more stably by the friction generated between the two.

On the other hand, in the first finger portion 60 _1, as shown in FIG. 13 (A) and (B), the fingertip portion 61 ₁ of the bones constituting the resin or the first consisting of a high hardness aluminum alloy material skeletal layer 70 _1, the Yubimoto unit 61 _first bone from the same material consisting second constituting skeletal layer 70 ₂ and has a first and second skeletal layer ₇₀ 1, 70 ₂ are first the first shaft 71 provided in parallel with the first finger portion 60 ₁ in the lateral direction (arrow a) constituting the knuckles 61 _3, and is integral with and bent freely connected.

Also at the lower end of the base of the finger portion 61 _2, provided as a second shaft 72 which constitutes the second finger joint 61 ₄ parallel to penetrate the base of the finger portion 61 ₂ and the first shaft 71 The second shaft body 72 is pivotally supported by a bearing 73 (FIG. 6A) provided at a predetermined position of the hand main body 62 (FIG. 6).

Thus in the first finger portion 60 _1, as described above with reference to FIG. 7, while maintaining the angle theta _th between the center axis _{k 2} of the third finger portion 60 _3, the fingertip portion 61 ₁ and the Yubimoto 61 ₂ can be independently bent to the inside 62A side of the hand portion.

Together this time, the first and second skeletal layer ₇₀ 1, 70 _2, as shown in FIG. 13 (C), alpha gel, the flexible layer 74 hardness such Sorubosein or urethane foam is made of a flexible material Hs0 And the flexible layer 74 is a surface layer formed of a flexible material such as rubber, PCV, or palliurethane having a thickness of about 0.3-1 [mm] and a hardness of about Hs 40-60. 75 is integrally covered.

Thus, in this robot 1, when the gripping such an object, a first resiliently and the inner surfaces of the fingers 60 ₁ in a direction that is recessed in accordance with the surface shape of the object is flexibly displaced the subject because it can be brought into close contact with the object, it is possible to reliably grip like the object, also the first finger portion 60 ₁ is the first and second skeletal layer 70 as described above _1, 70 _2, flexible Since it has a three-layer structure of the layer 74 and the surface layer 75, the surface of the first finger portion 601 is not easily damaged while giving flexibility to the surface of the _first finger portion 601, and can sufficiently withstand long-term use practically. It is made like that.

Similarly, in the fingers ₆₀ 2-60 ₅ of the second to fifth, as shown in FIG. 14 (A) and (B), the hardness of the resin or an aluminum alloy or the like constituting the bone of the fingertip portion 63 ₁ first skeletal layer 80 ₁ made of high material, and the second skeletal layer 80 ₂ of the same material which constitutes the bone of the finger central 63 _2, first made of the same material constituting the bone Yubimoto 63 ₂ 3 and a skeleton layer _803, these first to third skeletal layer ₈₀ 1-80 _3, the first shaft member 81 which constitutes the first knuckle 63 _3, second finger joint by a second shaft 82 which constitutes a 63 ₅ are integrally and bendably connected.

Also at the lower end of the base of the finger portion 63 _4, the third knuckle 63 ₆ third shaft 83 is the base of the finger portion 63 ₄ of which constitutes the parallel to the first and second shafts 81 and 82 The third shaft body 83 is provided so as to pass through, and is supported by corresponding bearings 84 _{1 to} 84 ₄ (FIG. 6A) provided at predetermined positions of the hand main body 62.

Thus in the finger portions ₆₀ 2-60 ₅ of the second to fifth, as described above for FIG. 6, the inner side of the fingertip portion 63 _1, finger middle 63 ₂ and Yubimoto 63 ₃ hand portion 23 to independently It can be bent freely toward

At this time, as shown in FIG. 14 (C), the _first to _third skeleton layers 80 ₁ to 80 ₃ are made of α gel, sorbosein, urethane foam, or the like, as in the case of the first finger portion 60 ₁ . The flexible layer 85 is integrally covered with a flexible layer 85 made of a flexible material having a hardness of Hs0, and the flexible layer 85 has a thickness of about 0.3 to 1 [mm] and a hardness of about Hs 40 to 60, such as rubber, PCV, or parylene urethane. It is integrally covered with a surface layer 86 formed using a flexible material.

Thereby, in this robot 1, when gripping an object, the inner surface of the _{second to fifth} finger portions 60 ₂ to 605 is elastic and flexible in a direction to dent according to the surface shape of the object. since by displacing can be brought into close contact with the object, the said object to be able reliably to grip the like, also second to fifth fingers 60 _{2-60 5} first through as described above Since the third skeleton layers 80 ₁ to 80 ₃ , the flexible layer 85 and the surface layer 86 have a three-layer structure, the surfaces of the _{second to fifth} finger portions 60 _{2 to} 605 have flexibility. However, the surface is not easily damaged, and can be sufficiently practically used even for long-term use.

In the case of this embodiment, as shown in FIG. 15, the tip portions of the fingertip portions 61 ₁ and 63 ₁ (FIGS. 13 and 14) in the first to fifth finger portions 60 _{1 to} 60 ₅ are the tip ends. The distal surfaces 70 ₁ A and 80 ₁ A of the first skeleton layers 70 ₁ and 80 ₁ that are formed so as to be curved so that the abdominal surfaces 60 ₁ A to 60 ₅ A approach the back surface as they go to , so as to approach the rear side of the finger portions ₆₀ 1 to 60 ₅ of the first to fifth as going to the first to fifth tip of the finger portions ₆₀ 1 to 60 _5, for example, 45 [°] of about tilt It is formed in a tapered shape having corners.

  Thereby, in the robot 1, the tip of the abdominal surfaces 601A to 605A in the fingertip portions 611 and 631 of the first to fifth finger portions 601 to 605 to which the force is most applied when the object is grasped is further increased. The contact area can be expanded by denting along the surface of the object, thus increasing the frictional force between the object and the first to fifth finger portions 601 to 605, and the object An object can be gripped more reliably.

In this embodiment, the surface layers 75 and 86 of the first to fifth finger portions 601 to 605 are hollow as shown in FIGS. 13C and 14C as a whole. And the abdominal surface shape of each of the fingertip portions 61 ₁ and 63 ₁ is selected to be a round square shape, whereby the first to fifth finger portions 60 _{1 to} 60 ₅ are defined as objects. The object can be reliably gripped by making contact with a wide contact area.

First to third fingers as in this case the first finger portion 60 ₁ is formed wider than the fingers ₆₀ 2-60 ₅ of the second to fifth, thereby for example, FIG. 11 (B) when picking an object 67 of paper or thin or the like by three-point support by the unit (60) _through 603 also to increase the contact area with respect to the first finger portion 60 ₁ of the object 67 to be 1-point support side The object 67 can be held more stably.

Further, in the case of this embodiment, on the side of the abdominal surface 60 ₁ A to 60 ₅ A of each fingertip portion 61 ₁ , 63 _{1 in} the _first to fifth finger portions 60 _{1 to} 60 ₅ , as shown in FIG. A fingerprint portion 87 having a concavo-convex pattern made up of a plurality of concentric concavo-convex portions is provided for improving the frictional force. As a result, for example, as shown in FIG. The second finger portions 601 and 602 can be picked up using friction generated between the finger portions 87 and the fingerprint portions 87.

Further, as shown in FIG. 16 (A), a hollow portion 87A having a predetermined depth is provided at the center of each fingerprint portion 87 in the _first to fifth finger portions 60 _{1 to} 60 ₅ , respectively. When the abdominal surfaces 60 ₁ A to 60 ₅ A of the _{first to fifth} finger portions 60 _{1 to} 60 ₅ are brought into contact with the object 90 as shown in FIG. 18, the first to fifth finger portions 60 ₁ to 60 ₁ to 60 ₅ air and air pressure rise in the recessed portion 87A in the fingerprint portion 87 from within the recess 87A of being pushed out, are adapted to give rise to adsorption in the recess portion 87A.

Thereby, in this robot 1, a target object can be hold | gripped more reliably, and, for example, the frictional force by each fingerprint part 87 of the _{1st- 5th} finger parts 601-605, and each said fingerprint part As shown in FIG. 19 (A), the ball 91 can be gripped at a position higher than the position of the center of gravity _GP by using the suction force in the hollow portion 87A of 87 and the flexibility of the fingertip. Has been made.

Further, in the case of this embodiment, the surface layers 75 and 86 of the _first to fifth finger portions 60 _{1 to} 60 ₅ are formed using a colorless and transparent material, and the surface layers 75 and 86 (FIG. 13 ( C), the boundary surface (the inner surface of the surface layers 75 and 86) with the flexible layers 74 and 85 (FIG. 13C and FIG. 14C) in FIG. 14C, or the surface layer in the flexible layers 74 and 85 Identification information unique to the robot 1 is printed as a two-dimensional barcode as shown in FIG. 20, for example, on the boundary surfaces with 75 and 86 (surfaces of the flexible layers 74 and 85).

  As a result, the robot 1 can be identified based on the two-dimensional barcode, thereby preventing confusion between the robots 1 and preventing the robot 1 from being stolen.

On the other hand, in the first to fifth finger portions 60 _{1 to} 60 ₅ , as shown in FIGS. 13 and 14, for example, the hardness of the resin material or the like is about Hs 70 and is slippery at the tip position on the back side. A claw 92 made of a material is attached to be exchangeable using an adhesive or the like.

In this case, the pawl 92 has its leading end portion is adhered to slightly protrude from the first to the fifth distal end of the finger portion 60 ₁ to 60 _5, are thereby placed on the desk surface or a floor surface A small object or paper can be picked by scratching the nail 92 protruding from the tips of the _first to fifth finger portions 60 _{1 to} 60 ₅ .

  Moreover, in the nail | claw 92, as evident also from FIG.13 and FIG.14, the front-end | tip part is formed in circular arc shape, and so that the nail | claw 92 can be equally scratched to a target object from each direction by this, In addition, when the nail 92 is scratched by an object, an external force acting on the nail 92 is dispersed to make the nail 92 difficult to break.

The tip of the nail 92 is formed so as to bend toward the abdominal surface 60 ₁ A to 60 ₅ A side of the _{first to fifth} finger portions 60 _{1 to} 60 ₅ , for example, as shown in FIG. 19 (B), and a small 93 first pawl 92 of the fingers 60 ₁ located on the floor, in the catch so by the second claw 92 of the fingers 60 _2, It is made so that it can be picked reliably.

(3) Drive Control of Hand 23 in Robot 1 (3-1) System Configuration Next, a system configuration of a drive control system that drives and controls the hand 23 having such a structure will be described.

In the case of this robot 1, the hand body portion 62 in the hand portion 23 is made to correspond to the _first to fifth finger portions 60 _{1 to} 60 ₅ as shown in FIGS. 21 (A) and 21 (B). The first to fifth actuator portions 100 _{1 to} 100 ₅ for driving the _first to fifth finger portions 60 _{1 to} 60 ₅ to extend or bend are housed, and the hand portion 23 and the forearm portion are stored. the wrist joint 101 for connecting the sixth actuator biaxial integral for driving the hand portion 23 in the roll direction (axial theta about the direction _R) and (around the axis theta _P) pitch direction A unit 102 is provided.

Also in the first to fifth actuator unit ₁₀₀ 1 to 100 _5, as shown in FIG. 22, the respective first to for controlling the actuator unit ₁₀₀ 1 to 100 ₅ of the first to fifth 5 The control devices 103 _{1 to} 103 ₅ are accommodated, and the sixth actuator unit 102 accommodates the sixth and seventh control devices 104 ₁ and 104 ₂ for the pitch axis and the roll axis. . Furthermore, an intelligent HUB 105 is housed in the actuator A ₈ (FIGS. 1 and 21) constituting the forearm portion of the robot 1, and the intelligent HUB 105 and the first to seventh control devices 103 _{1 to} 103 ₅ , 104 ₁ , 104 ₂ and are daisy chained.

In this case, as shown in FIG. 22, the intelligent HUB 105 includes a USB (Universal Serial Bus) interface circuit 110, a CPU (Central Processing Unit) 111, a serial communication interface circuit 112, a ROM (Read Only Memory), and a RAM (Random And a memory 113 such as an access memory, and communicates with the host controller (sub control unit 43C (FIG. 4)) via the USB interface circuit 110, and the first to first via the serial communication interface circuit 112. The seventh control devices 103 _{1 to} 103 ₅ , 104 ₁ and 104 ₂ can be serially communicated.

The intelligent HUB 105 communicates with the host controller at a cycle of 8 [ms], and the position command values (U1 (N)) from the host controller to the first to seventh control devices 103 _{1 to} 103 ₅ , 104 ₁ , and 104 ₂ . When various commands such as .about.U7 (N)) are given, servo commands (Ref1 (k) to Ref7 (k) having a period of 2 [ms] based on these position command values (U1 (N) to U7 (N)). )) Is generated and sent to the _first control device 1031 at the subsequent stage connected in a daisy chain.

At this time, when a servo command is given from the preceding intelligent HUB 105, the _first control device 1031 takes in the servo command (Ref1 (k)) for the self, and thereafter, the first control device 1031 corresponds to the first based on this servo command. while controlling the actuator unit 100 ₁ of 1, of the servo command given from the preceding intelligent HUB105 (Ref1 (k) ~Ref7 ( k)), a servo command to the self (Ref1 (k)), the first replacing the current position data representing the current position of the corresponding DC motor (described later) of the actuator portion 100 in the _first (P1 (k)), and sends them to the second control unit 103 ₂ of the subsequent stage.

Similarly, the second to seventh control devices 103 _{2 to} 103 ₅ , 104 ₁ , and 104 ₂ receive current position data and servos from the other first to sixth control devices 103 _{1 to} 103 ₅ , 104 _{1 in} the preceding stage. When the command is given, the servo commands (Ref2 (k) to Ref7 (k)) for the self are taken in, and thereafter, the corresponding _second to sixth actuator sections 100 _{2 to} 100 ₅ based on the servo commands are taken. , 101, while the servo commands and the current position data given from the _first to sixth control devices 103 _{1 to} 103 ₅ , 104 ₁ in the previous stage are assigned the corresponding second to second servo commands. 6 to current position data (P2 (k) to P7 (k)) representing the current position of a DC motor to be described later in the actuator sections 100 _{2 to} 100 ₅ , 104 ₁ and 104 ₂ . Instead, these are further sent to the other third to seventh control devices 103 _{3 to} 103 ₅ , 104 ₁ , 104 ₂ or the intelligent HUB 105 in the subsequent stage.

As a result, the intelligent HUB105 the current position data is given for each DC motors from the control unit 104 ₂ of the last 7, thus intelligent HUB105 is higher these current position data of each DC motor 8 [ms] cycle Send to controller. Based on the current position data of each of the DC motors, the host controller then creates new position command values (U1 (N) ˜) for the _first to seventh control devices 103 _{1 to} 103 ₅ , 104 ₁ and 104 ₂ . U7 (N)) is generated and given to the intelligent HUB 105 in the next cycle.

  In this way, in the robot 1, the host controller can perform the appearance control of the hand portion 23 via the intelligent HUB 105.

(3-2) Configuration of First to Sixth Actuator Parts 100 _{1 to} 100 ₅ , 102 Here, in the first to fifth actuators 100 _{1 to} 100 ₅ , as shown in FIG. The DC motor 121 is fixed as a power source. A gear 122 is attached to the output shaft 121 </ b> A of the DC motor 121, and the gear 122 meshes with a gear 123 formed of, for example, a crown gear in the gear box 120.

  Further, the gear 123 is fixed to a rotating shaft 124 that passes through the gear 123 coaxially, and the rotating shaft 124 is pivotally supported in the gear box 120 and is provided at both ends of the rotating shaft 124. The gripping rotary shaft 125 and the opening rotary shaft 126 are fixed coaxially.

And the one end side of the 1st wire 127 is wound in the direction which can be wound up when the DC motor 121 carries out normal rotation drive, and the front-end | tip part is being fixed to the surrounding side surface of the rotating shaft 125 for clamping. . The other end of the first wire 127, the corresponding first to fifth through the interior ventral side of the finger portions ₆₀ 1 to 60 ₅ of the first to fifth fingers ₆₀ 1 to 60 ₅ The end is fixed to the inner tip.

Thus, in the robot 1, the DC motor 121 of the _first to _fifth actuator units 100 _{1 to} 100 ₅ is driven to rotate forward so that the first wire 127 is wound on the peripheral side surface of the clamping shaft 125. the fingertip portion ₆₁ 1, 63 ₁ by the first wire 127 and to pull on the ventral side, can be driven so as to bend the fingers ₆₀ 1 to 60 ₅ of the corresponding first to fifth It is made like that.

In addition, one end portion of the second wire 128 is wound around the peripheral side surface of the opening rotation shaft 126 in a direction that can be wound when the DC motor 121 is driven in reverse rotation, and the tip end portion is fixed. The other end of the second wire 128, the corresponding first to fifth through the interior back side of the finger portions ₆₀ 1 to 60 ₅ of the first to fifth fingers ₆₀ 1 to 60 ₅ The end is fixed to the inner tip.

Thereby, in the robot 1, the DC motor 121 of the _first to _fifth actuator units 100 _{1 to} 100 ₅ is driven in reverse to wind the second wire 128 around the opening rotation shaft 126. By pulling the fingertip portions 61 ₁ and 63 ₁ to the back side by the second wire 128, the corresponding first to fifth finger portions 100 _{1 to} 100 ₅ can be driven to extend. ing.

  Further, an annular resin magnet 131 constituting the position detection sensor 130 is fixed to the end surface of the opening rotation shaft 126 coaxially with the opening rotation shaft 126 and is opposed to the opening rotation shaft 126 in parallel with a small distance. The control board 132 is arranged so that the first and second Hall elements 133A and 133B constituting the position detection sensor 130 together with the resin magnet 131 are mounted at predetermined positions facing the resin magnet 131 on the control board 132. Yes.

  At this time, the resin magnet 131 is magnetized in a predetermined pattern so that the magnetic pole density detected by the first and second Hall elements 133A and 133B changes according to the rotational position thereof. The current position of the corresponding DC motor 121 is detected based on first and second sensor signals (hereinafter referred to as first and second position detection signals) output from the two Hall elements 133A and 133B, respectively. It is made to be able to do.

Further, the control board 132 is formed with the first to fifth control devices 103 _{1 to} 103 ₅ described above with reference to FIG. 22 for driving and controlling the corresponding DC motor 121. Then, this is the first to fifth control unit ₁₀₃ 1 to 103 _5, servo command from the intelligent HUB105 as described above (Ref1 (k) ~Ref7 (k )) (FIG. 22), the above-described position detection First and second position detection signals output from the first and second Hall elements 133A and 133B of the sensor 130 (FIG. 23B) are provided. Thus, the first to fifth control devices 103 _{1 to} 103 ₅ drive and control the corresponding DC motor 121 based on the servo command and the first and second position detection signals.

In the case of this embodiment, the first and second wires 127, 128 are always wound around the rotary shaft 125 for clamping or the rotary shaft 126 for opening with a constant tension. When the external force in the bendable direction is applied to the fifth finger portions 60 _{1 to} 60 ₅ , the gripping rotary shaft 125 or the opening rotary shaft 126 is rotated accordingly. . As a result, in the robot 1, the first to fifth control devices 103 _{1 to} 103 ₅ correspond to the position detection sensors that the external force is applied to the first to fifth finger portions 60 _{1 to} 60 _5. Based on the first and second position detection signals from 130, it can be easily recognized.

  On the other hand, as shown in FIG. 24, the sixth actuator 102 has a roll axis DC motor 141 and a pitch axis DC motor 142 as power sources fixed at predetermined positions inside the housing 140. ing.

  In this case, the gear 143 is fixed to the output shaft 141A of the DC motor 141 for the roll shaft, and the gear 143 includes the gear 144, the gear 145 integrally formed with the gear 144, the gear 146, and the gear 146. The roll shaft output shaft 151 is connected through a gear 147, a gear 148, and a gear 149 and a gear 150, which are integrally formed with the gear 148, sequentially.

  At this time, as apparent from FIG. 21, the roll shaft output shaft 151 is fixed to the forearm portion of the robot 1, and thus the DC motor 141 is driven to rotate the hand portion 23 in the roll direction as a whole. Has been made to be able to.

  A gear 152 is fixed to the output shaft 142A of the DC motor 142 for the pitch axis, and the gear 152 is integrally formed with the gear 153, the gear 154 formed integrally with the gear 153, the gear 155, and the gear 155. The pitch shaft output shaft 158 is connected to the pitch shaft 158 and the gear 157 sequentially.

  At this time, the pitch shaft output shaft 158 is fixedly held by the first bearing 158A fixed to the rear end of the hand portion main body 62 of the hand portion 23, as is apparent from FIG. A rotating shaft 160 is rotatably provided coaxially with the pitch axis output shaft 158 on the surface of the housing 140 facing the pitch axis output shaft 158. The rotation shaft 160 is also shown in FIG. As is obvious, the second hand 159B fixed to the rear end of the hand portion main body 62 of the hand portion 23 is fixed and held.

  As a result, in the robot 1, the hand portion 23 can be rotationally driven in the pitch direction as a whole by rotationally driving the DC motor 142 in the sixth actuator portion 102.

  Further, in the sixth actuator portion 102, an annular resin magnet 162 is fixed coaxially with the roll shaft output shaft 151 on the inner surface side of the roll shaft output shaft 151, and in parallel with this, a slight distance is provided. The control board 163 is arranged so as to face each other, and the first and second Hall elements 164A and 164B constituting the position detection sensor 161 together with the resin magnet 162 are arranged at predetermined positions facing the resin magnet 162 on the control board 163. It is installed.

  At this time, like the resin magnet 131 described above with reference to FIG. 23, the resin magnet 162 has a predetermined pattern so that the magnetic pole density detected by the first and second Hall elements 164A and 164B changes according to the rotational position. The first and second sensor signals (hereinafter referred to as first and second position detection signals) output from the first and second Hall elements 164A and 164B, respectively, are magnetized. Based on this, the current position of the corresponding DC motor 141 can be detected.

The control board 163 includes a control unit 104 ₁ of the sixth described above with reference to FIG. 22 are formed. The sixth to the control device 104 _1, the servo command from the intelligent HUB105 as described above (Ref1 (k) ~Ref7 (k )) ( FIG. 22), the position detection sensor 161 described above (FIG. 24 (A )) And the first and second position detection signals outputted from the first and second Hall elements 164A and 164B. Thus the control unit 104 ₁ of the sixth, such a servo command, have been made to the DC motor 141 corresponding to control driving based on the first and second position detection signal.

  Similarly, an annular resin magnet 165 is fixed coaxially to the pitch shaft output shaft 158 on the inner surface side of the pitch shaft output shaft 158 and is opposed to the pitch shaft output shaft 158 in parallel with a small distance. The control board 166 is arranged so that the first and second Hall elements 168A and 168B constituting the position detection sensor 167 together with the resin magnet 165 are mounted at predetermined positions facing the resin magnet 165 on the control board 166. Yes.

  At this time, like the resin magnet 131 described above with reference to FIG. 23, the resin magnet 165 has a predetermined pattern so that the magnetic pole density detected by the first and second Hall elements 168A and 168B changes according to the rotational position. Magnetized so that the current position of the corresponding DC motor 142 can be detected based on the first and second position detection signals output from the first and second Hall elements 168A and 168B, respectively. Has been made.

The control board 166, the control unit 104 ₂ of the 7 described above is formed on FIG. And this is the seventh control unit 104 _2, a servo command from the intelligent HUB105 as described above (Ref1 (k) ~Ref7 (k )) ( FIG. 22), the position detection sensor 167 described above (FIG. 24 (A )) And the first and second position detection signals output from the first and second Hall elements. Thus the control unit 104 ₂ of the seventh, such a servo command, have been made to the DC motor 142 corresponding to control driving based on the first and second position detection signal.

(3-3) first to seventh control unit ₁₀₃ 1 to 103 ₅ of ₁₀₄ 1, 104 ₂ of the configuration where the first to seventh control unit ₁₀₃ 1 to 103 _{5, 104} 1, at 104 _2, As shown in FIG. 25, a motor control LSI (Large Scale Integrated circuit) 170 in which various communication and control circuits are formed on control boards 132, 163, and 166, and first and second Hall elements 133A, 133B, 164A, 164B, 168A, 168B are mounted, and a motor drive circuit 171 is formed on the control boards 132, 163, 166.

In this case, in the first to seventh control devices 103 _{1 to} 103 ₅ , 104 ₁ , 104 ₂ , the above-mentioned first output from the first and second Hall elements 133A, 133B, 164A, 164B, 168A, 168B. The first and second position detection signals S1A, S1B and the drive current detection signal S2 representing the coil current values of the corresponding DC motors 121, 141, 142 detected by the motor drive circuit 171 as described later are the motor control LSI 170. Given to.

Further, in the first to seventh control devices 103 _{1 to} 103 ₅ , 104 ₁ and 104 ₂ , the power supply line L _Vcc and the ground line L _GND, and two serial communication lines L _RXD for transmission and reception, Other first to seventh control devices 103 _{1 to} 103 ₅ , 104 ₁ , 104 ₂ or intelligent HUB 105 (FIG. 22) through cable 172 having L _TXD and one signal line L _CLK for clock input Thus, the motor control LSI 170 inputs drive power via this cable 172, and intelligently via the other first to seventh control devices 103 _{1 to} 103 ₅ , 104 ₁ , 104 ₂ etc. Communication with the HUB 105 is made possible.

  Then, the motor control LSI 170 transmits servo commands (Ref1 (k) to Ref7 (k)) (FIG. 22) given from the intelligent HUB 105 via the cable 172 and the like, and the first and second Hall elements 133A, 133B, Based on the first and second position detection signals S1A and S1B from 164A, 164B, 168A and 168B, and the drive current detection signal S2 from the motor drive circuit 171, it is applied to the DC motors 121, 141 and 142. The power drive current value is calculated, and a PWM (Pulse Width Modulation) signal S3 generated based on the calculation result is output to the motor drive circuit 171.

Thus, the motor drive circuit 171 applies the corresponding value of the drive current _ID to the corresponding DC motors 121, 141, 142 based on the PWM signal S 3 given from the motor control LSI 170, thereby causing the DC motors 121, 141. , 142 are driven.

  At this time, the motor drive circuit 171 detects the actual value of the drive current flowing through the coils of the DC motors 121, 141, 142, and the detection result is used as the drive current detection signal S2 as described above to the motor control system LSI 170. To send.

In this way, in the robot 1, servo commands given from the host controller by the first to seventh control devices 103 _{1 to} 103 ₅ , 104 ₁ , and 104 ₂ including the motor control LSI 170 and the motor drive circuit 171. Corresponding DC motors 121, 141, 142 can be driven accordingly.

(3-4) Configuration of Motor Control LSI 170 and Motor Drive Circuit 171 In the motor control LSI 170, as shown in FIG. 26, an arithmetic processing block 180, a register 181, and a subtractor 182 that constitutes a current control unit 189. , A current proportional gain multiplier 183, a PWM conversion block 184, a position detection block 185, and first and second analog / digital conversion circuits 186 and 187.

  In this motor control LSI 170, the first and second position detection signals S1A, S1B supplied from the first and second Hall elements 133A, 133B, 164A, 164B, 168A, 168B (FIG. 25) are supplied to the first and second position detection signals S1A, S1B. The first and second position detection data D1A and D1B obtained by digital conversion in the two analog / digital conversion circuits 188 are supplied to the position detection block 185.

The position detection block 185, based on the supplied first and second position detection data D1A and D1B, the DC motors 121 (in the corresponding first to fifth actuator units 100 _{1 to} 100 ₅ (FIG. 23)) ( 23) or the rotational position of the corresponding DC motor 141, 142 (FIG. 24) in the sixth actuator unit 102 (FIG. 24) is detected, and the current position data (P1 (P1 ( k) to P7 (k)) (FIG. 22) are stored in the register 181.

The arithmetic processing block 180 reads the current position data stored in the register 181 and reads the current position data thus read and servo commands (Ref1 (k) to Ref7 (k)) (FIG. 22) given from the intelligent HUB 105. On the basis of the target value (hereinafter referred to as a target value) of current necessary for generating torque for rotating the output shaft of the corresponding DC motor 121, 141, 142 (FIG. 25) to a specified rotation angle. I _{0 (} referred to as target current value) is calculated, and the obtained target current value I ₀ is stored in the register 181.

On the other hand, the target current value I ₀ stored in the register 181 is read by the subtractor 182. At this time, the subtractor 182 is supplied with drive current detection data D3 obtained by digitally converting the drive current detection signal S2 given from the motor drive circuit 171 (FIG. 25) in the first analog / digital conversion circuit 186. .

Thus the subtractor 182, the detected value of the actual coil current of the DC motor 121,141,142 at that time from the target current value _{I 0} is obtained based on the drive current detection data D3 is subtracted, resulting target current value I _The difference current value data D4 representing the difference with respect to ₀ is sent to the current proportional gain multiplier 183.

The current proportional gain multiplier 183 multiplies the difference current value data D4 by a predetermined current proportional gain Gp for converging the difference to “0”, and the DC motors 121, 141, 142 thus obtained are actually obtained. Is sent to the PWM conversion block 184 as a drive current target value I ₁ (hereinafter referred to as an applied current target value).

PWM conversion block 184, the applied current target value _{I 1} supplied to the PWM modulation, a PWM signal S3 obtained is sent to the motor driving circuit 171 as described above. At this time, the PWM conversion block 184 and the motor drive circuit 171 are connected by the first and second signal lines, and the PWM conversion block 184 performs PWM when driving the corresponding DC motors 121, 141, 142 in the normal direction. The signal S3 is sent to the motor drive circuit 171 via the first signal line, and a signal of logic "1" level (hereinafter referred to as a reference signal) S4 in the PWM signal S3 is sent to the other second signal line. To the motor drive circuit 171.

  The PWM conversion block 184 sends the PWM signal S3 to the motor drive circuit 171 via the second signal line and drives the reference signal to the first signal line when the corresponding DC motors 121, 141, 142 are driven in reverse rotation. To the motor drive circuit 171.

In the motor drive circuit 171, as shown in FIG. 27, a gate drive circuit 191 composed of two amplifiers 190A and 190B, two MOS (Metal Oxide Semiconductor) type N-channel FET (Field Effect Transistor) 192 ₁ , 192 ₂ and two MOS type P-channel FETs 192 ₃ and 192 ₄ are connected to each other in an inverter circuit 193 connected in a bridge shape.

Then, in the motor driving circuit 171, and the gate of the first signal line is the second MOS type P-channel FET192 _4, the first MOS type N-channel of the first inverter circuit 193 through the amplifier 190A of the gate drive circuit 191 The second signal line is connected to the gate of the FET 192 ₁ and the second signal line is connected to the gate of the first MOS type P-channel FET 192 ₃ and the second amplifier 190 B of the gate drive circuit 191. They are connected respectively to the gate of the MOS type N-channel FET192 _2.

Further, in this motor drive circuit 171, the connection midpoint of the first MOS type N-channel FET 192 ₁ and the first MOS type P-channel FET 192 ₃ , the second MOS type N-channel FET 192 ₂ and the second MOS type P-channel FET192 and _fourth connection midpoint is connected to the coil in the DC motor 121,141,142 corresponding respectively.

  Thereby, in the motor drive circuit 171, the PWM signal S3 given from the PWM conversion block via the first or second signal line is converted into an analog waveform drive current in the inverter circuit 193, and this is converted into the corresponding DC motor. The coils 121, 141, and 142 can be supplied as drive currents.

In the motor drive circuit 171, the first and second current detection terminals are connected between the source and the ground of the first MOS type P-channel FET 192 ₃ and between the source and the ground of the second MOS type P-channel FET 192 ₄ , respectively. Chip resistors R1 and R2 are connected.

  Then, the connection midpoint of the first current detection chip resistor R1 and the first MOS P-channel FET 1923, and the connection midpoint of the second current detection chip resistor R2 and the second MOS P-channel FET 1924, Are connected to the motor control LSI 170 (FIG. 26) via the differential amplifier 194, and the coil current detection results of the corresponding DC motors 121, 141, 142 detected by the motor drive circuit 171 are thereby described. Thus, the drive current detection signal S2 can be supplied to the motor control LSI 170.

(3-5) Specific Configuration of Each Processing Block of Motor Control LSI 170 Next, the configuration of the arithmetic processing block 180, position detection block 185, and PWM conversion block 184 of the motor control LSI 170 will be described in detail.

(3-5-1) Detailed Configuration of Arithmetic Processing Block 180 In the arithmetic processing block 180, as shown in FIG. 26, a CPU 200, a ROM 201 storing various programs, a RAM 202 as a work memory of the CPU 200, and a general purpose Between the parallel communication input / output circuit 203 corresponding to the parallel communication and other daisy chain connected first to seventh control devices 100 _{1 to} 103 ₅ , 104 ₁ , 104 ₂ (FIG. 22), etc. An input / output circuit 204 for serial communication, which is an input / output interface circuit, and a counter that generates a servo interrupt signal S10 with a period of 1 [ms] for a servo interrupt and a pulse signal S11 with a period of 50 [μs] as a PWM period. Servo interrupt signal S from timer / control circuit 205 and counter / timer / control circuit 205 A watchdog signal generation circuit 206 that generates a watchdog signal S12, which is a reference signal having a predetermined period of 1 [ms] period or more, for allowing the CPU 200 to determine whether or not 10 is generated correctly is mutually connected via the CPU bus 207. It is comprised by connecting to.

Then, the CPU 200 supplies power supply voltage Vcc (from the other first to seventh control devices 100 _{1 to} 103 ₅ , 104 ₁ , 104 ₂ (FIG. 22) etc. connected in a daisy chain via the serial communication input / output circuit 204. When FIG. 27) is supplied, first, based on an initial program stored in the ROM 201, a rising process such as an initial setting process for various control gains is executed.

  The CPU 200 thereafter performs a motor rotation control process for controlling the rotation of the corresponding DC motor 121, 141, 142 based on the servo interrupt signal S10 given from the counter / timer / control circuit 205 as a result. Run every [ms].

(3-5-2) Detailed Configuration of Position Detection Block 185 Next, the configuration of the position detection block 185 will be described in detail. As a premise thereof, the configuration of the position detection sensors 130 (FIG. 23) in the _first to _fifth actuator units 100 _{1 to} 100 ₅ (FIG. 23) and the position detection sensors 161 and 167 (FIG. 24) in the sixth actuator unit 102. Will be described.

  In each of the position detection sensors 130, 161, 167, as shown in FIG. 28, the resin magnets 131, 162, 165 have a magnetic flux density φ (θm) of the following formula along the circumferential direction.

It is magnetized so as to change. In this equation (1), φ ₀ represents the maximum magnetic flux density, and θ _m represents the rotation angle from the reference position P ₀ where the magnetic flux density is zero.

  On the other hand, the first and second Hall elements 133A, 133B, 164A, 164B, 168A, and 168B (FIGS. 23 and 24) of the position detection sensors 130, 161, and 167 are respectively shown in FIG. , 162, 165 and the control boards 132, 163, 166 (FIGS. 23, 24) so that the rotation angle is shifted by π / 2 concentrically with the resin magnets 131, 162, 165. It is mounted on.

The first and second position detection signals S1A and S1B output from the first and second Hall elements 133A, 133B, 164A, 164B, 168A, and 168B arranged in this way are shown in FIG. as shown, the sensor gain _{G 0,} the first and second Hall elements 133A, 133B, 164A, 164B, 168A, the maximum magnetic flux density was phi ₁ in the position of 168B, by using the above theta _m, respectively following formula

It can be expressed as

Thus, the first and second position detection signal S1A, based on S1B, it is possible to obtain the rotation angle theta _m of the resin magnet 131,162,165 by the following procedure.

That is, first, the rotation angle calculation value θ _x is set with the initial value set to 0, and the following formula

Is calculated.

If Eθ _x = 0, θ _x is

Calculated by Here, Kp represents a proportional gain, and Ki represents an integral gain, both of which are positive constants.

Using the calculated θ _x , the equation (4) is calculated again, and thereafter this is repeated until Eθ _x = 0. As a result, Eθ _x converges to a zero value. At this time, θ _x is expressed by the following equation:

As given.

Therefore, based on the equation (6) and θ _x calculated by the equation (5), the following equation:

Thus, the rotation angle θ _m of the resin magnet can be obtained. In the formulas (6) and (7), N represents an integer of 0 or more.

  Based on this principle, the position detection block 185 is configured as shown in FIG. 31, and the first and second Hall elements 133A, 133B, 164A, 164B, 168A, 168B of the position detection sensors 130, 161, 167 are configured. The first and second position detection signals D1A and D1B obtained by digitally converting the first and second position detection signals S1A and S1B output from the digital signal by the second analog / digital conversion circuit 187, respectively. 2 is input to the multipliers 210A and 210B.

At this time, the first and second multipliers 210A and 210B have the cosine value (cos θ _x ) or sine value (sin θ _x ) of the rotation angle calculation value θ _x calculated in advance as described below, respectively. Given from the second function converters 214A and 214B.

Thus, the first multiplier 210A multiplies the first position detection data D1A and the cosine value (cos θ _x ) of the rotation angle calculation value θ _x and sends the multiplication result to the subtractor 211. The second multiplier 210 _{</ b} > B multiplies the second position detection data D _<b > 1 _{</ b} > B by the rotation angle calculation value θ _x and the sine value (sin θ _x ), and sends the multiplication result to the subtractor 211.

  The subtractor 211 obtains the operation result given by the equation (4) by subtracting the multiplication result of the second multiplier 210B from the supplied multiplication result of the first multiplier 210A, and obtains the result of the third multiplication. To the device 212.

  Then, the multiplication result is obtained from the following expression in the third multiplier 212 using S as a Laplace operator:

Are sequentially multiplied by an integral gain and a proportional gain Kr, and then multiplied by 1 / S (S is a Laplace operator) in a fourth multiplier 213.

As a result, a rotation angle calculation value θ _x is obtained as a multiplication result in the fourth multiplier 213, and this is given to the first and second function calculators 214A and 214B. The first and second function calculators 214A and 214B respectively calculate the sine value (sin θ _x ) and cosine value (cos θ _x ) of the supplied rotation angle calculation value θ _x , and the calculation result as described above. This is applied to the second or first multiplier 210A, 210B.

On the other hand, the rotation angle calculation value θ _x output from the fourth multiplier 213 is stored in the register 181 (FIG. 26). And the rotation angle calculated value theta _x, thereafter read out by the processing block 180, the current position data representing the current position of the DC motor 121,141,142 corresponding to the above (P1 (k) ~P7 ( k)) is used to control the DC motors 121, 141, 142.

Note arithmetic processing of the rotation angle calculation value theta _x as described above in the position detecting block 185 is performed based on the pulse signal S11 supplied from the counter timer control circuit 205 of the operation processing block 180. Thus, the rotation angle calculation value theta _x output from the position detecting block 185, and thus to be updated is the period of the pulse signal S11 every 50 [μs].

(3-5-3) Detailed Configuration of PWM Conversion Block 184 The PWM conversion block 184 controls the pulse width of a 50 [μs] cycle pulse based on the multiplication result supplied from the current proportional gain multiplier 183. As a result, the PWM modulation signal S3 and the reference signal S4 are generated and sent to the motor drive circuit 171 (FIG. 25).

In practice, as shown in FIG. 31, the PWM conversion block 184 sets the applied current target value I ₁ (FIG. 26) supplied from the current proportional gain multiplier 183 in an internal register (not shown) and applies the applied current target value I. when ₁ is positive, given the applied current target value I ₁ from the counter timer control circuit 205 of the operation processing block 180 (FIG. 26), first for each rising edge of the 50 [μs] period of the PWM pulse signal S11 1 is set in a down counter (not shown) in the PWM pulse signal generation circuit 220A.

  The down counter decreases the counter value at each rising edge of the CPU clock (0.1 [μs]) of the arithmetic processing block 180 and stops at the zero value. Therefore, the output of the first PWM pulse signal generation circuit 220A is the logic “1” level until the count value of the down counter reaches zero, and the logic “0” level after the counter value reaches zero. Become.

The process described above is repeated applied current target value I ₁ stored in the re-register on the rising edge of the next PWM pulse signal S11 is again set to the down counter of the first PWM pulse signal generation circuit 220A.

Therefore, from the first PWM pulse signal generation circuit 220A, to applied current target value I ₁ stored in the register is updated, a predetermined pulse width T _on of the PWM signal proportional to the applied current target value I ₁ S3 is output, and the second PWM pulse signal generation circuit 220B outputs a reference signal S4 having a logic “0” level.

On the other hand, in the PWM conversion block 184, after conversion to a positive integer by computing the absolute value when the applied current target value I ₁ is negative, the applied current target value I ₁ second PWM A down counter (not shown) in the pulse signal generation circuit 220B is set.

The result this time, from the second PWM pulse signal generating circuit 220B, as in the first PWM pulse signal generation circuit 220A described above, until the applied current target value I ₁ stored in the register is updated, PWM signal S3 constant pulse width _{T on} proportional to the applied current target value _{I 1} is output. At this time, the first PWM signal generation circuit 220A outputs a reference signal S4 having a logic “1” level.

In this way, in the PWM conversion block 184, and generates a PWM signal S3 and the reference signal S4 having a pulse width corresponding to the applied current target value I ₁ of the calculation result supplied from the current proportional gain multiplier 183, a motor this It can be sent to the drive circuit 171.

(4) Software Control in Arithmetic Processing Block 180 (4-1) Series of Software Control Flow Next, software control in the arithmetic processing block 180 will be described.

  The CPU 200 (FIG. 26) of the arithmetic processing block 180 executes various processes for driving and controlling the corresponding DC motors 121, 141, 142 in accordance with the data processing procedure shown in FIG.

  That is, the CPU 200 executes this data processing procedure every time the servo interrupt signal S10 is given from the counter / timer / control circuit 205 (FIG. 26), and first controls the serial communication input / output circuit 204 (FIG. 26). As a result, the servo commands (Ref1 (k) to Ref7 (k)) (FIG. 22) transmitted from the intelligent HUB 105 (FIG. 22) and various other commands are received, and these are received in the RAM 202 (FIG. 26). Store (step SP1).

  Next, the CPU 200 performs predetermined data reception processing such as data format conversion processing on the servo commands and various commands stored in the RAM 202 (step SP2). Incidentally, communication with the intelligent HUB 105 (FIG. 22) is performed at a cycle of 2 [ms], and therefore, the servo command from the intelligent HUB 105 is given every 2 [ms], whereas the counter / timer / control circuit 205 Since the servo interrupt signal S10 is given every 1 [ms], the reception processing in step SP1 and step SP2 as described above is performed once every two times of execution of this data processing procedure. Become.

  Subsequently, the CPU 200 corresponds to the servo command (Ref1 (k) to Ref7 (k)) obtained every 2 [ms] obtained by the data reception process, and the previously obtained corresponding stored in the RAM 202. Based on the current position data (P1 (k) to P7 (k)) (FIG. 22) of the DC motors 121, 141, 142, the target position for each [ms] of the DC motors 121, 141, 142 is expressed. Servo commands (Ref1 (k) ′ to Ref7 (k) ′) are generated (step SP3).

  Note that the CPU 200 reads a system initial setting value stored in advance in the ROM 201 (FIG. 26) instead of the servo command at the initial stage when the servo command or the like is not yet given from the host controller such as when the power is turned on. Similar processing is executed based on the system setting value.

Further, the CPU 200 replaces the servo command with the current proportional gain _GP described above for the current proportional gain multiplier 183 (FIG. 26) from the host controller or the expression (2) of the position detection block 185 (FIG. 26). When a command for changing the control gain such as the sensor gain G ₀ and the compliance adjustment gain Kadj related to the compliance control described later is given, the corresponding control gain stored in the RAM 202 is changed according to the command.

On the other hand, when the CPU 200 generates the servo commands (Ref1 (k) ′ to Ref7 (k) ′) every 1 [ms] as described above, the servo commands (Ref1 (k) ′ to Ref7 (k) ′). ) And current position data (P1 (k) to P7 (k)) given from the position detection block 185 at this time, rotation control processing for controlling the rotation of the corresponding DC motors 121, 141, 142 Execute. Specifically, the CPU 200 performs the rotation control process based on servo commands (Ref1 (k) ′ to Ref7 (k) ′) and current position data (P1 (k) to P7 (k)). The current command value I ₀ described above with reference to FIG. 26 is calculated and stored in the register 181 (step SP4).

In addition, the CPU 200 also stores the current position data from the position detection block 185 and internal variable data such as the difference e ₁ (FIG. 34) obtained in the compliance control processing described later (hereinafter referred to as motor internal variable data). Stored in the RAM 202 (FIG. 26).

  The CPU 200 then executes predetermined data transmission processing such as data conversion processing for transmitting the current position data of the corresponding DC motors 121, 141, 142 stored in the RAM 202, motor internal variable data, and the like to the outside. The obtained current position data and motor internal variable data after the data transmission processing are stored in the RAM 202 (step SP5).

Then, the CPU 200 controls the serial communication input / output circuit 204 (FIG. 26) thereafter, so that the current position data and the motor internal variable data after the data transmission process stored in the RAM 202 are daisy chained. The second to seventh control devices 103 _{2 to} 103 ₇ , 104 ₁ , 104 ₂ or the intelligent HUB 105 are transmitted (step SP6).

  In this way, in the arithmetic processing block 180, various corresponding processes are executed based on the servo commands and commands from the intelligent HUB 105.

(4-2) Specific Contents of Rotation Control Processing Here, regarding the rotation control processing at step SP4 of the data processing procedure described above with reference to FIG. 33, the CPU 200 in the _first to fifth control devices 103 _{1 to} 103 ₅ Specific processing contents will be described.

During the rotation control process, the CPU 200 performs servo commands (Ref1 (k) ′ to Ref5 (k) ′) for every 1 ms generated based on the servo commands (Ref1 (k) to Ref5 (k)) from the intelligent HUB 105. ) And the current position data (P1 (k) to P5 (k)) detected by the position detection block 185, the position control process and the compliance control process are simultaneously executed in parallel to achieve the target current value. to generate a I _0.

  Incidentally, the position control refers to a target rotation angle that is a servo command (Ref1 (k) ′ to Ref5 (k) ′) every 1 [ms], and current position data (P1 (k) to P5 (k)). ) Based on the difference between the current rotation angle (current position) of the output shaft of the DC motor 121 (FIG. 23) obtained on the basis of the difference and controlling the drive current applied to the corresponding DC motor 121, Control that converges the difference to “0”.

The compliance control here controls the amount of displacement of the DC motor 121 so that when the external force is applied to the DC motor 121, the magnitude of the external force is proportional to the amount of displacement of the DC motor 121. That means. The proportionality constant at this time is referred to as a compliance adjustment gain, and when the compliance adjustment gain is “0”, even if there is an external force, the displacement amount is “0”, that is, the rigidity is high. Further, as the compliance adjustment gain becomes larger than “0”, the displacement amount gradually increases and the rigidity becomes low. In this way, the amount of displacement due to the external force of the DC motor 121 can be controlled in accordance with the magnitude of the compliance adjustment gain. For example, the external force is applied to the first to fifth finger portions 60 _{1 to} 60 ₅ (FIG. 6) of the robot 1. Sometimes, the external force can prevent the first to fifth finger portions 60 _{1 to} 60 _{5 from} being damaged, and the softness when the robot 1 grips the object can be controlled. A method for controlling such softness is compliance control.

  The CPU 200 executes the position control process and the compliance control process simultaneously in parallel, thereby performing the appearance control of the hand portion 23 with rigidity according to a preset compliance adjustment gain.

In practice, the CPU 200 performs the servo control (Ref1 (k) ′ to Ref5 (k) ′) every 1 [ms] and the position at this time as shown in FIG. A difference e ₁ from the current position Yf of the DC motor 121 given from the detection block 185 is calculated (step SP10), and the corresponding DC motor 121 is actually driven by multiplying the difference e ₁ by a position proportional gain Kp. A target current value I ₀ ′ for control is calculated (step SP11).

Then, the CPU 200 calculates the target current value I ₀ by adding the target current value I ₀ ′ and the compliance adjustment value Tref described later (step SP12), and stores this in the register 181 (FIG. 26) as described above. Store. As a result, the target current value I ₀ is read by the subtractor 182 (FIG. 26), and thereafter, based on the target current value I ₀ , the position as described above is set so that the difference e1 is “0”. Control is performed.

  On the other hand, the CPU 200 calculates the current position of the DC motor 121 based on a mathematical model having the same characteristics as that of the DC motor 121 simultaneously and in parallel (step SP13).

Specifically, the CPU 200 calculates the servo current command (Ref1 (k) ′ to Ref5 (k) ′) every 1 [ms] and the current theoretical position of the corresponding DC motor 121 calculated by the mathematical model. By calculating a difference e ₂ from YMf (hereinafter referred to as the model current position) (step SP13A) and multiplying the difference e ₂ by a position proportional gain KMp having the same value as the position proportional gain Kp of the actual model. Then, a theoretical target current value I _{0 (m)} ′ corresponding to the target current value I ₀ ′ of the actual model (hereinafter referred to as a model target current value) I _{0 (m)} ′ is calculated (step SP13B).

Then, the CPU 200 sets the difference current in the current proportional gain multiplier 183 described above with reference to FIG. 26 as a current proportional gain for converging the difference e ₂ to “0” with respect to the model target current value I _{0 (m)} ′. by multiplying the current proportional gain GMp the same value as the current proportional gain Gp to be multiplied by the value data D4, the theoretical applied current target value corresponding to the application of the actual model current target value I _{1 (FIG.} 26) (hereinafter, This is called a model applied current target value) I _{1 (m)} (step SP13C).

Subsequently, the CPU 200 multiplies the model applied current target value I _{1 (m)} by the motor constant model KMm having the same value as the motor constant Km of the DC motor 121 corresponding to the actual model, thereby obtaining A theoretical output torque TMm of the DC motor 121 corresponding to the output torque Tm of the DC motor 121 (hereinafter referred to as model output torque TMm) is calculated (step SP13D).

  Further, the CPU 200 thereafter adds the model output torque TMm and the external force estimated torque Tf_est, which is an estimated external force torque calculated as described later, to the output shaft of the DC motor 121 in the actual model at this time. A total torque Tf, which is a theoretical total of the torques generated, and a theoretical total torque (hereinafter referred to as model total torque) TMf corresponding to the total torque Tf are calculated (step SP13E).

  The CPU 200 then calculates the following equation for the model total torque TMf:

The model current position YMf of the corresponding DC motor 121 described above is calculated by multiplying a constant corresponding to the mechanical characteristics of the first to fifth fingers 60 _{1 to} 60 ₅ on the mathematical model given by (Step SP13F). In this equation (9), JMy and DMy have the same values as the mechanical moment of inertia Jy or the friction coefficient Dy of the _first to _fifth actuator units 100 _{1 to} 100 ₅ (FIG. 23) in the actual model, respectively. It is a constant and S represents a Laplace operator.

  Next, the CPU 200 subtracts the model current position YMf from the current position (hereinafter referred to as the actual model current position) Yf of the DC motor 121 in the actual model detected by the position detection block 185, thereby obtaining these actual models. A difference Yfer between the current position Yf and the model current position YMf is calculated (step SP14).

Further, the CPU 200 multiplies the difference Yfer by a predetermined estimated gain Kest to obtain the above-described external force estimated torque Tf_est that is the estimated torque of the external force applied to the corresponding first to fifth finger portions 60 _{1 to} 60 _5. calculate.

Here, the when the external force estimated torque Tf_est adds a model target current value I ₀ as described above _{(m) ',} the model current position YMf changes like the actual model current position Yf calculated by mathematical model, the actual The difference Yfer between the model current position Yf and the model current position YMf converges to “0”.

Since the estimated external force torque Tf_est when the difference Yfer converges to “0” becomes the same value as the external force torque Tf_ext that has actually acted on the first to fifth finger portions 60 _{1 to} 60 ₅ . This external force torque Tf_ext can be obtained with high accuracy as the external force estimated torque Tf_est.

  The convergence response time at this time is determined by the magnitude of the estimated gain Kest. The larger the estimated gain Kest, the shorter the convergence response time, and the smaller the estimated gain Kest, the longer the convergence response time.

On the other hand, the CPU 200 calculates the compliance adjustment value Tref by multiplying the external force estimated torque Tf_est calculated in this way by the above-described compliance adjustment gain Kadj, and the position control processing as described above. Is added to the target current value I ₀ ′ calculated by the above.

As a result, the corresponding DC motor 121 responds separately from the position command values U1 (N) to U5 (N) from the main control unit 40 (servo command values Ref1 (k) to Ref5 (k) from the intelligent HUB 105). The first to fifth finger portions 60 _{1 to} 60 ₅ are driven in accordance with the magnitude of the external force acting on the _first to fifth finger portions 60 _{1 to} 60 ₅ and the magnitude of the compliance adjustment gain Kadj set at that time. when an external force is applied to the finger portions ₆₀ 1 to 60 ₅ of 5, finger portions ₆₀ 1 to 60 ₅ of the first to fifth is to bend in the same direction as the external force in accordance with the external force.

  In this case, the value of the compliance adjustment gain Kadj is

As a result, since the same torque as the external force torque Tf_ext is added as the compliance adjustment value Tref, the corresponding movements of the _first to fifth finger portions 60 _{1 to} 60 ₅ react with the external force by the position control process. The force is canceled. Thus, at this time, the first to fifth finger portions 60 _{1 to} 60 ₅ can be bent lightly by an external force.

  On the other hand, the value of the compliance adjustment gain Kadj is

When this is the case, the corresponding movements of the _first to fifth finger portions 60 _{1 to} 60 ₅ are movements in which the reaction force against the external force by the position control process as described above is attenuated. Thus, at this time, the first to fifth finger portions 60 _{1 to} 60 ₅ can be bent with a reaction force having a magnitude corresponding to the compliance adjustment gain Kadj with respect to the external force.

Thus, by adjusting the value of the compliance adjustment gain Kadj, the reaction force of the _first to fifth finger portions 60 _{1 to} 60 ₅ against the external force can be controlled. Then Such compliance control, when an external force is exerted on the finger portions ₆₀ 1 to 60 ₅ of the first to fifth, the first to fifth fingers ₆₀ 1 to 60 ₅ is the upper-level controller (main controller 40 (FIG. 5)), it moves according to the external force against the position command values U1 (N) to U5 (N) from the first to fifth fingers resulting from the action of the external force. The breakage of 60 _{1 to} 60 ₅ can be prevented in advance and effectively, and an external object can be softly gripped along its shape.

(5) Compliance Control in Robot 1 (5-1) Series of Flows Related to Compliance Control Next, a series of flows related to actual compliance control in the robot 1 will be described with reference to FIG.

In this robot 1, under the control of the main controller 40, the compliance adjustment gain Kadj in the corresponding first to fifth control devices 103 _{1 to} 103 ₅ is changed according to the situation, so that While the first to fifth finger portions 60 _{1 to} 60 ₅ are given high rigidity, the rigidity of the first to fifth finger portions 60 _{1 to} 60 ₅ is lowered when gripping an external object or when the robot 1 falls. The external object is softly gripped along its shape, and can be effectively prevented from being damaged due to the first to fifth finger portions 60 _{1 to} 60 ₅ colliding with the floor surface. Yes.

Further, in this robot 1, when an unexpected external force is applied to the first to fifth finger portions 60 _{1 to} 60 ₅ , the control of the main control unit 40 is left and the first to fifth corresponding to the intelligent HUB 105 corresponds. By changing the compliance adjustment gain Kadj in the control devices 103 _{1 to} 103 ₅ , the first to fifth finger portions 60 _{1 to} 60 ₅ are made soft against the external force in a reflective manner, and the external force is applied. The first to fifth finger portions 60 _{1 to} 60 ₅ due to the above can be effectively prevented from being damaged.

In other words, in the case of this robot 1, the calculation unit 40B of the main control unit 40 first to fifth fingers stored in the external memory 58 (FIG. 5) in advance when the robot 1 is powered on. The normal values of compliance adjustment gain Kadj for normal time (hereinafter referred to as normal value of normal adjustment gain) for units 60 _{1 to} 60 ₅ are read out and sent to intelligent HUB 105.

In addition, the calculation unit 40B of the main control unit 40 thereafter positions relative to the first to fifth control devices 103 _{1 to} 103 ₅ based on the next action content determined by the autonomous control process as described above. The command values U1 (N) to U5 (N) are sequentially generated and sent to the intelligent HUB 105 via the sub-control unit 43C corresponding to each communication period of 8 [ms].

The intelligent HUB 105 manages the configuration (thumb, index finger, middle finger, ring finger, and little finger) of the first to fifth control devices 103 _{1 to} 103 _{5 in} the component ratio adjustment unit 105B, and is transmitted from the main control unit 40. The normal adjustment gain prescribed values for the first to _fifth finger portions 60 _{1 to} 60 ₅ are transmitted to the _first to fifth control devices 103 _{1 to} 103 ₅ . Thus, the first to fifth control devices 103 _{1 to} 103 ₅ take in corresponding ones of the normal-time adjustment gain prescribed values, and set the values as initial values of the compliance adjustment gain Kadj.

Further, the intelligent HUB 105 thereafter performs servo command values Ref1 (k) to 2 [ms] based on the position command values U1 (N) to U5 (N) transmitted from the main control unit 40 at intervals of 8 [ms]. Ref5 (k) is generated and transmitted to the _first to fifth control devices 103 _{1 to} 103 ₅ every 2 [ms] communication period. Thus, the first to fifth control devices 103 _{1 to} 103 ₅ thereafter transmit the servo command values Ref1 (k) to Ref5 (k) transmitted from the intelligent HUB 105 and the compliance adjustment gain Kadj set as described above. Based on the above, the position control process and the compliance control process are simultaneously executed in parallel.

On the other hand, the calculation unit 40B of the main control unit 40 determines to hold the external object as the next action, and when actually causing the robot 1 to hold the external object, the calculation unit 40B stores the external object in the external memory 58 (FIG. 5) in advance. The stored prescribed values of the compliance adjustment gain Kadj for gripping of the _first to fifth finger portions 60 _{1 to} 60 ₅ (hereinafter referred to as “adjustment gain defining values for gripping”) are read out.

In this case, the gripping adjustment gain prescribed values are selected to be larger than the corresponding normal adjustment gain prescribed values. Thus, the gripping adjustment gain prescribed values are used as the compliance adjustment gain Kadj. By setting the _fifth control devices 103 _{1 to} 103 ₅ respectively, the first to fifth finger portions 60 _{1 to} 60 ₅ can be made to be less rigid than normal.

The calculation unit 40B of the main control unit transmits the read regulation value for adjustment gain at the time of the _first to fifth finger portions 60 _{1 to} 60 ₅ to the intelligent HUB 105 via the corresponding sub control unit 43C. To do.

Further, the calculation unit 40B of the main control unit 40 uses that the gravity detected by the acceleration sensor 56 becomes “0” when the robot 1 is in a fall state (a state from when the robot 1 starts to fall until it finishes falling). When the fall state of the robot 1 is detected based on the acceleration detection signal S2B supplied from the acceleration sensor 56, the first to fifth finger portions 60 _{1 to} 60 ₅ stored in the external memory 58 in advance are detected. The prescribed values of the compliance adjustment gain Kadj for the fall (hereinafter referred to as the fall adjustment gain prescribed values) are read out and sent to the intelligent HUB 105. The fall adjustment gain prescribed values are also selected to be larger than the corresponding normal adjustment gain prescribed values. Then, the calculation unit 40B transmits these to the intelligent HUB 105 via the corresponding sub control unit 43C.

At this time, the intelligent HUB 105 sequentially generates servo command values Ref1 (k) to Ref5 (k) every 2 [ms] based on the position command values U1 (N) to U5 (N) transmitted from the main control unit 40. To do. The intelligent HUB 105 transmits the servo command values Ref1 (k) to Ref5 (k) and the compliance adjustment gain Kadj transmitted from the main control unit 40 to the first to fifth control devices 103 _{1 to} 103 ₅ .

The first to fifth control devices 103 _{1 to} 103 ₅ control the rotation of the corresponding DC motor 121 based on the transmitted servo command values Ref1 (k) to Ref5 (k), while adjusting the gripping adjustment gain. Value or the fall adjustment gain specified value is transmitted, the setting value of the compliance adjustment gain Kadj that has been set in the self until then is updated to that value, and based on the updated new compliance adjustment gain Kadj The compliance control for the corresponding first to fifth finger portions 60 _{1 to} 60 ₅ is executed.

As a result, when the external object is gripped, the first to fifth fingers 60 ₁ to 60 ₁ to 60 ₁ to 60 ₁ so that the reaction force (external force) received from the external object always has a constant magnitude corresponding to the gripping adjustment gain specified value. 60 _{for 5} operates, it is gripped soften the external object by bending the fingers 60 ₁ to 60 ₅ of the first to fifth in imitation even complicated shapes external object to the shape at constant pressure Can do. Also, when the robot 1 falls and the first to fifth finger portions 60 _{1 to} 60 ₅ collide with the floor surface, the first to fifth finger portions 60 _{1 to} 60 ₅ have low rigidity. The first to fifth finger portions 60 _{1 to} 60 ₅ bend in the direction of the collision force according to the collision force (external force) applied to the first to fifth finger portions 60 _{1 to} 60 ₅ from the floor surface, It is possible to effectively prevent the _first to fifth finger portions 60 _{1 to} 60 _{5 from} being damaged by the collision force.

On the other hand, the first to fifth control devices 103 _{1 to} 103 ₅ correspond to the current positions (P1 (k) to P5 (k)) of the corresponding DC motor 121 detected by the position detection block 185 (FIG. 26), The external force estimated torque Tf_est calculated by the method described above for 34 is transmitted to the intelligent HUB 105 as a state quantity for the corresponding first to fifth finger portions 60 _{1 to} 60 ₅ every 2 [ms] communication cycle.

The intelligent HUB 105 transmits the state quantity transmitted from the _first to fifth control devices 103 _{1 to} 103 ₅ to the main control unit 40 via the sub control unit 43C corresponding to each communication period of 8 [ms]. On the other hand, the value of the external force estimated torque Tf_est for the _first to fifth finger portions 60 _{1 to} 60 ₅ among these state quantities is constantly monitored by the calculation unit 105A.

When the value of the external force estimated torque Tf_est for the _first to fifth finger portions 60 _{1 to} 60 ₅ becomes equal to or greater than a predetermined threshold stored in advance, the arithmetic unit 105A of the intelligent HUB 105 stores in advance the memory 113 (FIG. 22). Reflection control for the corresponding first to fifth control devices 103 _{1 to} 103 ₅ among the reflection control adjustment gain prescribed values for the _first to fifth finger portions 60 _{1 to} 60 ₅ stored in The adjustment gain specified value is read out and transmitted to the _first to fifth control devices 103 _{1 to} 103 ₅ .

Thus, at this time, the first to fifth control devices 103 _{1 to} 103 ₅ update the setting value of the compliance adjustment gain Kadj to the reflection control adjustment gain specified value, and respond based on the updated compliance adjustment gain Kadj. The compliance control for the _first to fifth finger portions 60 _{1 to} 60 ₅ is executed.

In this case, the reflection control adjustment gain prescribed values for the _first to fifth fingers 60 _{1 to} 60 ₅ are selected to be larger than the corresponding normal adjustment gain prescribed values. Accordingly, since the first to fifth finger portions 60 _{1 to} 60 ₅ are softer than the normal force at this time, the external force acts on the _first to fifth finger portions 60 _{1 to} 60 _5. It is possible to effectively prevent damage due to the failure.

(5-2) Specific Processing Contents of Each Unit Here, the main control unit 40 executes various processes related to compliance control as described above according to the first compliance control processing procedure RT1 shown in FIG.

That is, when the power of the robot 1 is turned on, the main control unit 40 starts the first compliance control processing procedure RT1 at step SP0, and is stored in the external memory 58 (FIG. 5) in advance at step SP1. 34, various control gains such as the compliance adjustment gain Kadj, the external force estimation gain Kest, the position proportional gains Kp and KMp, and the current proportional gain GMp described above with reference to FIG. Various machine constants such as the coefficient DMy and the configurations (thumb, index finger, middle finger, ring finger, and little finger) of the first to fifth control devices 103 _{1 to} 103 ₅ are read out and correspond to the sub-control unit 43C (FIG. 4). , FIG. 5) to the intelligent HUB 105 .

  Further, when the main control unit 40 intends to hold an external object as the next action of the robot 1 in the subsequent step SP2, based on the image signal S1A (FIG. 5) given from the CCD camera 50 (FIG. 5), for example, The distance to the external object is detected by a stereo ranging method or the like. In step SP3, the main control unit 40 detects the posture of the robot 1 based on the acceleration detection signal S2B (FIG. 5) supplied from the acceleration sensor 56 (FIG. 5).

Then, the main control unit 40 proceeds to step SP4, and based on the detection result in step SP2, determines whether or not the external object to be gripped is within a grippable distance at that time, and obtains a positive result. Then, proceeding to step SP6, the adjustment gain regulation values for grasping for the _first to fifth finger portions 60 _{1 to} 60 ₅ are read from the external memory 58 (FIG. 5), respectively, and the corresponding sub control unit 43C is read out. To the intelligent HUB 105, and then proceeds to step SP8. Note that step SP2 and step SP4 are omitted when there is no plan to hold an external object.

  On the other hand, when step SP4 or the like is omitted or when a negative result is obtained in step SP4, the main control unit 40 proceeds to step SP5, and based on the detection result in step SP3, the robot 1 is currently Judge whether it is in a fall state.

If the main control unit 40 obtains an affirmative result in step SP5, the main control unit 40 proceeds to step SP6 to adjust for the _first to fifth finger portions 60 _{1 to} 60 ₅ from the external memory 58 (FIG. 5). Each of the specified gain values is read out and sent to the intelligent HUB 105 via the corresponding sub-control unit 43C before proceeding to step SP8.

On the other hand, if the main control unit 40 obtains a negative result in step SP5, the main control unit 40 proceeds to step SP7, and the normal-time adjustment gain prescribed value (for example, "" for the _first to fifth finger portions 60 _{1 to} 60 _5). 0 ") are read out and sent to the intelligent HUB 105 via the corresponding sub-control unit 43C, and then the process proceeds to step SP8. Note that when the normal adjustment gain prescribed value is already set as the compliance adjustment gain Kadj in the _first to fifth control devices 103 _{1 to} 103 ₃ , this step SP7 is omitted.

When the main control unit 40 proceeds to step SP8, the position command value U1 (k) to be transmitted to the _first to fifth control devices 103 _{1 to} 103 ₅ according to the action content to be executed by the robot 1 at that time. .About.U5 (k) (FIG. 22) is calculated and sent to the intelligent HUB 105 via the corresponding sub-control unit 43C.

  Further, the main control unit 40 thereafter returns to step SP2, and thereafter repeats the same processing for step SP2 to step SP8 at a cycle of 8 [ms].

In this way, the main control unit 40 changes the compliance adjustment gain Kadj in the _first to fifth control devices 103 _{1 to} 103 ₅ according to the situation.

  On the other hand, the CPU 111 (FIG. 22) of the intelligent HUB 105 executes various processes related to compliance control as described above in accordance with the second compliance control processing procedure RT2 shown in FIG.

That is, when the power of the robot 1 is turned on, the CPU 111 starts the second compliance control processing procedure RT2 in step SP10, and in the subsequent step SP11, the main control unit 40 controls the above-described various control gains, various machine constants, and the like. The initial value of compliance adjustment gain Kadj (ordinary adjustment gain standard value) and the configurations of first to fifth control devices 103 _{1 to} 103 ₅ (thumb, index finger, middle finger, ring finger, and little finger) are transmitted. Await. Upon receiving these, the CPU 111 transmits various control gains, various machine constants, and initial values of the compliance adjustment gain Kadj to the corresponding first to fifth control devices 103 _{1 to} 103 ₅ , while The configurations of the fifth to third control devices 103 _{1 to} 103 ₅ are managed by the configuration ratio adjusting unit 105B (FIG. 35).

  Then, the CPU 111 proceeds to step SP12 to change the position command values U1 (k) to U5 (k) and the change value of the compliance adjustment gain Kadj (normal adjustment gain specified value, adjustment for gripping) from the main control unit 40. If a specified gain value or a fall adjustment gain specified value) is transmitted, the process proceeds to step SP13. Further, when these have not been transmitted, the CPU 111 proceeds to step SP13 as it is.

Then, when the CPU 111 proceeds to step SP13, the servo for the _first to fifth control devices 103 _{1 to} 103 ₅ is based on the last received position command values U1 (k) to U5 (k) from the main control unit 40. Command values Ref1 (k) to Ref5 (k) are generated and transmitted to the corresponding first to fifth control devices 103 _{1 to} 103 ₅ , respectively, while the changed value of the compliance adjustment gain Kadj is received in step SP12. When this is done, this is transmitted to the corresponding first to fifth control devices 103 _{1 to} 103 ₅ .

Further, the CPU 111 thereafter proceeds to step SP14, and the current positions (P1 (k) to P5 (P) of the corresponding DC motors 121 (FIG. 23) transmitted from the _first to fifth control devices 103 _{1 to} 103 ₅ respectively. k)) and external force estimated torque Tf_est (FIG. 34), and then proceeds to step SP15, where any of the received external force estimated torques Tf_est for the _first to fifth finger portions 60 _{1 to} 60 _{5 is} received. Is determined to be greater than a corresponding threshold value set in advance for each of the _first to fifth finger portions 60 _{1 to} 60 ₅ .

If the CPU 111 obtains a negative result in step SP15, it returns to step SP12, and if it obtains an affirmative result, it proceeds to step SP16, and the first to fifth finger portions 60 _{1 to} 60 ₅ are preliminarily determined. The predetermined adjustment gain regulation value for reflection control is read from the memory 113 (FIG. 22) and transmitted to the corresponding first to fifth control devices 103 _{1 to} 103 ₅ .

  Further, the CPU 111 thereafter returns to step SP12, and thereafter repeats the same processing for step SP12 to step SP16 at a cycle of 2 [ms].

In this way, the CPU 111 of the intelligent HUB 105 performs compliance in the _first to fifth control devices 103 _{1 to} 103 ₅ when an external force of a certain level or more is applied to the first to fifth finger portions 60 _{1 to} 60 _5. The adjustment gain Kadj is changed.

On the other hand, the CPU 200 (FIG. 26) of the _first to fifth control devices 103 _{1 to} 103 ₅ executes various processes related to compliance control as described above in accordance with the third compliance control processing procedure RT3 shown in FIG.

  That is, when the power of the robot 1 is turned on, the CPU 200 starts the third compliance control processing procedure RT3 in step SP20, and in the subsequent step SP21, various control gains, various mechanical constants, and compliance adjustment gains from the intelligent HUB 105. Wait for Kadj to be sent. When receiving these, the CPU 200 sets them as the initial values of the corresponding control gain, mechanical constant, or compliance adjustment gain Kadj.

Then, the CPU 200 proceeds to step SP22 to calculate the external force estimated torque Tf_est for the corresponding first to fifth finger portions 60 _{1 to} 60 ₅ , and then proceeds to step SP23 to be transmitted from the intelligent HUB 105. The servo command values Ref1 (k) to Ref5 (k) and the change value of the compliance adjustment gain Kadj are received. Then, the CPU 200 thereafter controls the rotation of the corresponding DC motor 121 based on the servo command values Ref1 (k) to Ref5 (k). On the other hand, when the change value of the compliance adjustment gain Kadj is transmitted, The set value of the compliance adjustment gain Kadj is changed to the changed value.

  Subsequently, the CPU 200 proceeds to step SP24, and transmits the external force estimated torque Tf_est calculated in step SP22 to the intelligent HUB 105. Thereafter, the CPU 200 returns to step SP22, and thereafter performs the same processing for steps SP22 to SP24 1 [ ms].

In this way, the CPUs 200 of the _first to fifth control devices 103 _{1 to} 103 ₅ correspond to the corresponding DC based on the servo command values Ref 1 (k) to Ref 5 (k) and the compliance adjustment gain Kadj transmitted from the intelligent HUB 105. The rotation of the motor 121 is controlled at 1 [ms].

(6) Operation and effect of the present embodiment In the above configuration, the robot 1 normally supplies the normal adjustment gain specified value as the compliance adjustment gain Kadj to the first to fifth control devices 103 _{1 to} 103 _5. set, while controlling the softness for the first to fifth fingers 60 ₁ to 60 ₅ of the external force on the basis of the normal time adjustment gain specified value, when fall of the grip or when the robot 1 of the external object, the The compliance adjustment gain Kadj set in the _first to fifth control devices 103 _{1 to} 103 ₅ is changed to a grip adjustment gain specified value or a fall adjustment gain specified value larger than the normal adjustment gain specified value. and, to control the softness for the grasping during adjustment gain specified value or the first to fifth force fingers 60 _through 603 ₅ based on the fall adjustment gain specified value.

Therefore, this robot 1, while in the normal remembering high rigidity fingers 60 ₁ to 60 ₅ of the first to fifth, gripping or during tipping the first to fifth fingers during 60 of the robot 1 of _the external object it is possible to lower the rigidity of 60 _5, while it is possible to perform the task of high rigidity fingers is required, soft, slightly modeled after the fingers the external object on the surface having a complex outer shape it is possible to have further also the first to break the fifth finger portion 60 ₁ to 60 ₅ in a fall can be effectively prevented.

In the robot 1, the compliance adjustment set in the _first to fifth control devices 103 _{1 to} 103 ₅ when an external force of a certain level or more is applied to the first to fifth finger portions 60 _{1 to} 60 _5. The set value of the gain Kadj is changed to the reflection control adjustment gain specified value under the control of the intelligent HUB 105.

Accordingly, in this robot 1, the rigidity of the first to fifth fingers ₆₀ 1 to 60 ₅ is reflectively lower when unexpected external force acts on the fingers ₆₀ 1 to 60 ₅ of the first to fifth Therefore, damage to the _first to fifth finger portions 60 _{1 to} 60 ₅ due to the external force can be effectively prevented.

According to the above configuration, the normal adjustment gain prescribed value is set as the compliance adjustment gain Kadj in the _first to fifth control devices 103 _{1 to} 103 ₅ at the normal time, and based on the normal adjustment gain prescribed value. While controlling the softness of the _first to fifth finger portions 60 _{1 to} 60 ₅ against the external force, the first to fifth control devices 103 _{1 to} 103 ₅ are used when holding an external object or when the robot 1 falls. The set compliance adjustment gain Kadj is changed to a gripping adjustment gain specified value or a fall adjustment gain specified value that is larger than the normal adjustment gain specified value. by based on use adjustment gain prescribed value so as to control the softness for the first to fifth force fingers 60 _through 603 _5, high rigidity fingers is required Work While it is possible to perform, it is possible to have soft, slightly modeled after the fingers the external object on the surface having a complex outer shape, further the first to fifth fingers 60 ₁ at a fall 60 ₅ of breakage can also be effectively prevented, thus possible to realize a first to fifth robot capable of improving the usability while effectively preventing damage to the fingers 60 ₁ to 60 _5.

(7) Other Embodiments In the above-described embodiment, the case where the present invention is applied to the biped walking humanoid robot 1 configured as shown in FIG. 1 is described. The present invention is not limited to this, and can be widely applied to robot apparatuses having various configurations.

Further, in the above-described embodiment, the compliance adjustment set in the _first to fifth control devices 103 _{1 to} 103 ₅ is performed only when the robot 1 grips an external object and when the robot 1 is in a falling state. The case where the gain Kadj is changed has been described, but the present invention is not limited to this, and the compliance adjustment gain Kadj set in the _first to fifth control devices 103 _{1 to} 103 ₅ is changed at other timings. In short, what is necessary is to change the compliance adjustment gain Kadj when necessary.

In this case, in the above-described embodiment, when the robot 1 grips an external object, the first to fifth control devices 103 _{1 to} 103 ₅ are always set regardless of the type of external object to be gripped. The case where the compliance adjustment gain Kadj is changed to a fixed gripping adjustment gain prescribed value has been described, but the present invention is not limited to this, and for example, a plurality of gripping adjustment gain prescribed values are prepared. When gripping an external object, the external object is identified by image recognition processing, and when the external object is fragile, such as an egg, the smallest adjustment gain regulation value for gripping is set as the compliance adjustment gain Kadj On the other hand, when the external object is hard and heavy, the compliance adjustment gain Kadj has the largest gripping time. And so setting the adjustment gain specified value, it may be used different as gripping during adjustment gain specified value in accordance with the grasped object.

Furthermore, in the above-described embodiment, the external force estimated torque Tf_est calculated by the CPU 200 of the _first to fifth control devices 103 _{1 to} 103 ₅ is monitored, and the value of the external force estimated torque Tf_est is set based on a preset threshold value. Is larger, the set value of the compliance adjustment gain Kadj set in the _first to fifth control devices 103 _{1 to} 103 ₅ so as to lower the rigidity of the corresponding first to fifth finger portions 601 to 605. Although the intelligent HUB 105 is provided with a function as an external force monitoring means for changing the control gain to the reflection control adjustment gain specified value, the present invention is not limited to this, and this function is not limited to the first to fifth control devices. 103 ₁ to 103 may be imparted to the ₅ side (e.g., CPU 200).

  The present invention can be widely applied to various types of robot apparatuses in addition to humanoid robots.

It is a perspective view which shows the external appearance structure of the robot by this Embodiment. It is a perspective view which shows the external appearance structure of the robot by this Embodiment. It is a perspective view with which it uses for description of the external appearance structure of the robot by this Embodiment. It is a block diagram with which it uses for description of the internal structure of a robot. It is a block diagram with which it uses for description of the internal structure of a robot. It is the rough-line front view and side view which show the external appearance structure of the hand part of a robot. It is the rough-line front view and side view with which it uses for description of a structure of a hand part. It is the rough-line front view and side view with which it uses for description of a structure of a hand part. It is the rough-line front view and side view with which it uses for description of a structure of a hand part. It is the rough-line front view and side view with which it uses for description of a structure of a hand part. It is the rough-line front view and side view with which it uses for description of a structure of a hand part. Schematic front view for explaining the hand main _body, A ₁ -A 1 'sectional _view, A ₂ -A 2' is a cross-sectional view. It is a rough-line side view, back view, and sectional drawing which show the structure of a 1st finger | toe part. It is an approximate line side view, back view, and sectional view showing composition of the 2nd-5th finger part. It is sectional drawing with which it uses for description of the structure of a fingertip. It is a rough-line side view and front view with which it uses for description of the shape of a fingertip. It is an approximate line side view used for explanation of operation which grasps paper. It is a rough-line side view with which it uses for description of a fingerprint part. It is an approximate line side view used for explanation of operation of grasping and picking. It is a basic diagram which shows a two-dimensional barcode. FIG. 2 is a schematic front view and a side view for explaining a system configuration of a hand drive control system. It is a block diagram with which it uses for description of the connection relationship of intelligent HUB and the 1st-7th control apparatus. It is the front view and side view which show the structure of a 1st-5th actuator part substantially linearly. It is the front view and side view which show the structure of a 6th actuator part substantially. It is a block diagram which shows the structure of the 1st-7th control apparatus. It is a block diagram which shows the structure of LSI for motor control. It is a circuit diagram which shows the structure of a motor drive circuit. It is a conceptual diagram with which it uses for description of the magnetization pattern of a resin magnet. It is a basic diagram which shows the positional relationship of a resin magnet and the 1st and 2nd Hall element. It is a wave form diagram with which it uses for description of the 1st and 2nd position detection signal. It is a block diagram which shows the structure of a position detection block. It is a conceptual diagram with which it uses for description of a PMW conversion block. It is a conceptual diagram with which it uses for description of the software control in an arithmetic processing block. It is a block diagram with which it uses for description of rotation control of the DC motor in an arithmetic processing block. It is a block diagram with which it uses for description of a series of flows regarding the compliance control of a robot. It is a flowchart which shows a 1st compliance control processing procedure. It is a flowchart which shows a 2nd compliance control processing procedure. It is a flowchart which shows the 3rd compliance control processing procedure.

Explanation of symbols

1 ...... robot, 40 ...... main control _unit, 601 through 603 ₅ ...... _fingers, 103 ₁ to 103 _5, 104 1, 104 ₂ ...... controller 105 ...... intelligent HUB, 111,200 ...... CPU 121, 141, 142 ... DC motor, 130, 161, 167 ... position detection sensor, 170 ... motor control LSI, 180 ... arithmetic processing block, 185 ... position detection block, S1A, S1B ... position Detection signal, U1 (N) to U7 (N) ... Position command value, Ref1 (k) to Ref7 (k), Ref1 (k) 'to Ref7 (k)' ... Servo command value, P1 (k) to P7 (k): Current position data.

Claims (8)

In a robot apparatus provided with a movable part that can be displaced,
A motor as a power source for driving the movable part;
A host controller that outputs a command according to the action content to be executed;
Control means for controlling the rotation of the motor in accordance with the command given from the host controller;
Sensor means for detecting the rotational state of the motor;
External force estimation means for estimating the magnitude of external force acting on the movable part based on the detection result of the sensor means,
The control means includes
Based on the estimation result of the external force estimating means, the motor has a proportional relationship in which the magnitude of the external force and the displacement amount of the rotation state of the motor due to the external force have a preset adjustment gain as a proportional constant. Control the rotation of
The upper controller is
The robot apparatus, wherein the adjustment gain set in the control means is changed according to a situation. The movable part is
A finger provided on the hand of the robot device;
The upper controller is
The robot apparatus according to claim 1, wherein when the robot apparatus grips an external object, the adjustment gain set in the control means is changed so as to reduce the rigidity of the finger unit. The movable part is
A finger provided on the hand of the robot device;
The upper controller is
2. The robot apparatus according to claim 1, wherein the adjustment gain set in the control unit is changed so that the rigidity of the finger portion is lowered when a fall of the robot apparatus is detected. Provided separately from the host controller, comprising external force monitoring means for monitoring the estimation result of the external force estimation means,
The external force monitoring means is
When the magnitude of the external force estimated by the external force estimation means is larger than a preset magnitude, the adjustment gain set in the control means is set so as to reduce the rigidity of the movable part with respect to the external force. The robot apparatus according to claim 1, wherein the robot apparatus is updated. In the control method of the robot apparatus provided with the movable unit that can be displaced,
A first step of controlling rotation of a motor as a power source for driving the movable part according to the action content to be executed, and detecting a rotation state of the motor;
A second step of estimating the magnitude of an external force acting on the movable part based on the detected rotational state of the motor;
Based on the estimation result of the magnitude of the external force, the magnitude of the external force and the amount of displacement in the rotational state of the motor due to the external force have a proportional relationship with a preset adjustment gain as a proportional constant. A third step of controlling the rotation of the motor;
And a fourth step of changing the adjustment gain set in the control means according to the situation. The movable part is
A finger provided on the hand of the robot device;
In the fourth step,
The robot apparatus according to claim 5, wherein when the robot apparatus grips an external object, the adjustment gain set in the control means is changed so as to reduce the rigidity of the finger portion. Control method. The movable part is
A finger provided on the hand of the robot device;
In the fourth step,
The control of the robot apparatus according to claim 5, wherein the adjustment gain set in the control means is changed so that the rigidity of the finger portion is lowered when the fall of the robot apparatus is detected. Method. In the second or third step,
Monitor the estimation result of the magnitude of the external force acting on the movable part,
In the fourth step,
When the estimated magnitude of the external force is larger than a preset magnitude, the adjustment gain set in the control means is updated so as to reduce the rigidity of the movable part with respect to the external force. The method for controlling a robot apparatus according to claim 5.

Images