JP2012045194A - Exoskeleton type robot - Google Patents

Exoskeleton type robot Download PDF

Info

Publication number
JP2012045194A
JP2012045194A JP2010190485A JP2010190485A JP2012045194A JP 2012045194 A JP2012045194 A JP 2012045194A JP 2010190485 A JP2010190485 A JP 2010190485A JP 2010190485 A JP2010190485 A JP 2010190485A JP 2012045194 A JP2012045194 A JP 2012045194A
Authority
JP
Japan
Prior art keywords
torque
robot
exoskeleton
control
joint
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
JP2010190485A
Other languages
Japanese (ja)
Inventor
Soko Gen
相昊 玄
Original Assignee
Advanced Telecommunication Research Institute International
株式会社国際電気通信基礎技術研究所
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Advanced Telecommunication Research Institute International, 株式会社国際電気通信基礎技術研究所 filed Critical Advanced Telecommunication Research Institute International
Priority to JP2010190485A priority Critical patent/JP2012045194A/en
Publication of JP2012045194A publication Critical patent/JP2012045194A/en
Pending legal-status Critical Current

Links

Abstract

PROBLEM TO BE SOLVED: To solve the problems of conventional exoskeleton type robots of incapability of properly conducting motion support to trunk and lower limb parts, and low energy efficiency in cases of using electric motors and hydraulic actuators that generate a large load, resulting in energy consumption by the actuators even in supporting the load only.SOLUTION: The exoskeleton type robot with a base and a lower half body includes active joints as active type joints disposed at respective positions of left and right ankles, left and right knees, and left and right hips, and a control part to actuate the active joints. The active joint includes air muscles and electric motors.

Description

  The present invention relates to an exoskeleton robot.
  In recent years, research on a brain machine interface (BMI) that controls external robot devices using brain information for the purpose of motion reconstruction has attracted attention (Non-Patent Document 1). On the other hand, robots that can balance and walk have been developed. For example, there is a robot that can optimally distribute an action force necessary for movement to a plurality of contact points in space and generate torque of each joint in the same way as a human (see Patent Document 1).
  In the conventional BMI research, most of them have provided communication means such as cursor control on a monitor and control of a robot arm, and aimed to reconstruct an upper limb motion. On the other hand, despite the potential demand, no BMI research has been conducted to support exercise support for the trunk and lower limbs. The difficulty of constructing BMI to support trunk and lower limb movements is that the brain's control mechanism for trunk and lower limb movements is not fully understood, and human ability to walk and adjust posture This is because it was not technically easy to construct an exercise support robot device having a performance close to that of the robot.
  Nevertheless, several exoskeleton-type robots have been developed to support lower limb and trunk movements. Non-patent documents 2, 3, and 4 are typical examples of exoskeleton robots. Among these, the type that supports the weight uses a large electric motor or hydraulic cylinder, and the weight of the robot body is considerably large. On the other hand, as rehabilitation applications that do not support weight, there are an increasing number of cases where pneumatic artificial muscles (hereinafter referred to as air muscles) are used (Non-Patent Document 5).
  Air muscle is lighter and cheaper to manufacture than other actuators, and has the unique air pressure softness. On the other hand, it has been said that it is not suitable for precise control because of delay, nonlinearity, and individual variation.
WO2007 / 139135 Publication
Lebedev, M.A. and Nicolelis, M.A .: `` Brain-machine interfaces past, present and future '', Trends in Neuroscience, vol. 29, no. 9, pp. 536-546, 2006. Jacobsen, S .: `` On the Development of XOS, a Powerful Exoskeletal Robot '', IEEE / RSJ IROS, Plenary Talk, 2007. Kazerooni, H., Chu, A., Steger, R .: `` That Which Does Not Stabilize, Will Only Make Us Stronger '', The International Journal of Robotics Research, vol.26, no.1, pp.75- 89,2007. Suzuki, K., Mito, G., Kawamoto, H., Hasegawa, Y., Sankai, Y.: `` Intension-based walking support for paraplegia patients with Robot Suit HAL '', Advanced Robotics, vol.21, no .12, pp.1441-1469,2007. Akio Nakagawa, et al .: "Rehabilitation support robot using pneumatic rubber artificial muscle", Fluid Power System, vol.38, no.4, pp.194-198, 2007. Sardellitti1, I., Park, J., Shin, D., Khatib, O., `` Air muscle controller design in the distributed Macro-Mini (DM2) actuation approach '', IEEE / RSJ IROS, pp. 1822-1827 , 2007. Satoshi Nakata, Yasuo Sakurai, Hiroshi Tanaka: `` Study on combined electro-pneumatic drive system and its control method '', Proceedings of Japan Fluid Power System Society, vol.39, no.1, 2008.
  Accordingly, an object of the present invention is to appropriately provide exercise support for the trunk and lower limbs by proposing an exoskeleton robot of a pneumatic and electric hybrid drive system.
  As mentioned above, the air muscle is lightweight, low friction, and can generate a large force, so it has an attractive point as an actuator for an exoskeleton robot to be worn on a person, but it has a large time delay and variation. Because it has non-linearity, it is not suitable for accurate and agile torque control. Therefore, Non-Patent Document 6 describes an attempt to improve the disadvantages of the air muscle by combining the air muscle and the electric motor.
  However, there is a problem in applying this as it is to an exoskeleton robot that supports exercise of the trunk and lower limbs. This is because, in the technique of Non-Patent Document 6, antagonistic driving is performed using two air muscles. In this case, the range of motion becomes very narrow due to the relationship between stroke and force, or a large torque is generated. This is because it cannot occur. In addition, the drive system itself which combines an electric motor with an air cylinder has already been studied in Non-Patent Document 7.
  Therefore, in the present invention, an exoskeleton robot that is light and can withstand practical use is realized by optimally arranging the air muscle in consideration of the dynamics of the robot and the output characteristics of the air muscle.
  That is, the present invention is a hybrid drive type exoskeleton robot that is not found in the prior art.
  The exoskeleton type robot according to the first aspect of the present invention is an exoskeleton type robot having a base and a lower body, with active joints arranged at left and right ankles, left and right knees, and left and right positions of the waist. The active joint is an exoskeleton type robot that includes an active muscle and an electric motor. The active joint includes an active joint and a control unit that operates the active joint. In this robot, considering the dynamics of the robot and the output characteristics of the air muscle, the air muscles are optimally arranged so that the air muscle can generate a large force when a large torque is required for the joint.
  With this configuration, torque control of the trunk and lower limbs can be performed appropriately. Appropriate is agility, high load, and high precision.
  Further, in the exoskeleton robot according to the second invention, the control unit operates the air muscle and the electric motor with respect to the torque having a frequency higher than the first threshold value. This is an exoskeleton-type robot that operates only the air muscle with respect to a torque having a frequency lower than the threshold value.
  With this configuration, torque control of the trunk and lower limbs can be performed appropriately.
  In the exoskeleton robot according to the third aspect of the present invention, in contrast to the first aspect, the control unit operates only the air muscle with respect to torque having a lower load than the second threshold, It is an exoskeleton type robot that operates an air muscle and an electric motor with respect to a torque that is higher than the load.
  With this configuration, torque control of the trunk and lower limbs can be performed appropriately.
  In the exoskeleton robot of the fourth invention, the control unit has a lower load than the second threshold and a lower frequency than the first threshold compared to the second or third invention. It is an exoskeleton robot that operates only the air muscle with respect to the torque, operates the air muscle and the electric motor with respect to the torque having a higher load than the second threshold and a higher frequency than the first threshold. .
  With this configuration, torque control of the trunk and lower limbs can be performed appropriately.
  According to the exoskeleton robot of the present invention, exercise support for the trunk and lower limbs can be performed appropriately. In addition, this robot is much lighter than other driving systems with the same output, thanks to the lightweight air muscle. When the air valve is closed, the spring characteristics of the air muscle can increase the load without energy consumption. Because it can be maintained, it is extremely energy efficient compared to other drive systems.
The figure which shows the model of the prototype of the exoskeleton type robot in Embodiment 1 Block diagram of the exoskeleton robot The figure which shows the freedom degree composition of the exoskeleton type robot Graph showing the characteristics of the hybrid drive Schematic diagram of the active joint of the hybrid drive Schematic diagram of the active joint of the hybrid drive Diagram showing the link mechanism used in the experiment Figure showing the results of the experiment Figure showing snapshot during the experiment Figure showing the experimental data Diagram showing comparison between the same hybrid drive and other systems
Hereinafter, embodiments of an exoskeleton type robot and the like will be described with reference to the drawings. In addition, since the component which attached | subjected the same code | symbol in embodiment performs the same operation | movement, description may be abbreviate | omitted again.
(Embodiment 1)
  In this embodiment, an aerodynamic hybrid exoskeleton robot for walking / posture rehabilitation will be described.
  This exoskeleton type robot has an exoskeleton. An exoskeleton is a skeletal structure. An exoskeleton-type robot has a base and a lower body, and has joints with 6 degrees of freedom in the left and right positions of the ankle, knee, and waist. Further, the six joints are aerodynamic hybrid drive joints.
  FIG. 1 is a diagram showing a configuration example of an exoskeleton robot 1 in the present embodiment. The exoskeleton robot 1 has 10 degrees of freedom. 1A and 1B are perspective views of the exoskeleton robot 1 viewed from different angles. In FIG. 1, the exoskeleton robot 1 includes a backpack 101, a flexible sheet 102, an HAA antagonist muscle 103, an HFE extensor muscle 104, an HFE motor 111, a KFE extensor muscle 105, a KFE motor 106, an AFE extensor / AAA antagonist muscle 107. AFE flexor muscle 108, universal joint 109, and rotary joint 110 with pulley.
  In addition, a wire encoder is attached to the universal joint 109 to measure the joint angle. A wire encoder is similarly attached to the rotary joint 110 with pulley. Note that in the HFE and KFE joints which are hybrid joints, the joint angle may be measured using an encoder attached to the motor.
  FIG. 2 shows a configuration diagram of the degree of freedom of the exoskeleton robot 1. In FIG. 1, only the HFE and KFE joints are hybrid driven, but it goes without saying that other configurations are possible, such as hybrid driving all. Also, in FIG. 1, out of all 10 degrees of freedom, the left and right HAA joints and AFE, and the AAA joint employs antagonistic driving by extensors and flexors. When the movable angle and the required torque are not large, it is possible to reduce the weight more than using the electric motor together. In FIG. 1, in particular, the AFE and AAA joints are movable by coordinating three air muscles that are smaller in size than the upper joints. In either case, the responsiveness is sacrificed because the electric motor is not used together.
  This will be supplemented below. In an exercise example in which overall performance is important, such as posture control, the following configuration can be used. For example, if the AFE joint is appropriately co-contracted, a sufficient balance can be obtained by precise and dynamic torque control of HFE and KFE. It should be noted that appropriately co-contracting the AFE joint is to contract the two antagonist muscles together. At this time, the joint has a spring-like characteristic. In fact, it has been demonstrated that dynamic balance is possible even when the AFE joint torque is zero.
  In FIG. 1, a posture sensor is mounted on the body portion to detect the posture of the base portion. In addition, wire encoders are attached to all joints so that joint angles can be measured. In this configuration example, it is assumed that the same method as that of Patent Document 1 is implemented as the whole body motion control algorithm. In this case, an accurate Jacobian matrix from the center of gravity to the contact portion can be calculated only by detecting the base posture and the joint angle, and the target torque to be generated at each joint can be calculated.
  A floor reaction force sensor is mounted on the sole. This is used in an auxiliary manner to determine whether or not the sole that assumes contact is actually in contact, and to correct a model error included in the Jacobian matrix. This method is essentially different from conventional control methods that rely entirely on feedback information from the floor reaction force sensor for attitude control. The conventional control method is position control type force feedback control.
  In addition to the controller, the backpack 101 incorporates an air muscle valve and an electric motor driver. In addition, it is equipped with a battery, a compressed CO2 gas cylinder, and a regulator, enabling short-term autonomous driving in case the power supply line and air supply are disconnected. Since the robot has a sufficient payload, a small pneumatic pump may be mounted here if noise is not a problem. In that case, a CO2 gas cylinder is unnecessary.
  FIG. 3 is an example of a block diagram of the exoskeleton robot 1.
  The exoskeleton robot 1 includes a reception unit 11, an exoskeleton 12, and a control unit 13. The exoskeleton 12 includes a base 121, a lower body 122, an active joint 123, and a detection mechanism 124. Further, the active joint 123 includes an air muscle 1231 and an electric motor 1232. The control unit 13 includes a recording unit 131, a storage unit 132, a measuring unit 133, a control unit 134, and an output unit 135.
  The accepting unit 11 can accept a torque command or a position command commanded from another computer or decoded from brain information. Here, “acceptance” is a concept that usually includes reception of information transmitted via a wired or wireless communication line, reception of information read from a recording medium such as an optical disk, a magnetic disk, or a semiconductor memory. . The brain information is, for example, information indicating that the right leg is raised, information indicating walking and walking speed, and information that regulates the movement of the trunk and lower limbs including the torque value of a specific joint. It is. Note that the reception unit 11 may be considered as a part of the control unit 13.
  The material which comprises the exoskeleton 12 is not ask | required.
  The base 121 may be considered to include the skeleton at the waist position and the active joint 123 at the waist position, or may be considered to be only the skeleton at the waist position.
  The lower body 122 may be considered to include the skeleton at the position of the thigh or foot, the active joint 123 at the position of the thigh or foot, or may be considered to be only the skeleton at the position of the thigh or foot.
  The active joints 123 are active joints arranged at left and right ankles, left and right knees, and left and right positions of the waist. Here, the active joint 123 is a joint that can be actively operated by an actuator. That is, the active joint 123 includes an actuator. Further, the one or more active joints 123 here are of a hybrid type. That is, the active joint 123 is a hybrid type that includes an air muscle 1231 and an electric motor 1232. Moreover, the active joint 123 here comprises 10 degrees of freedom, for example. The actuator has a function of receiving a torque value as a control target value as a drive signal and controlling based on the received torque value. The actuator may be of any type, such as a servo motor or a hydraulic motor. For example, in a servo motor that has a drive circuit capable of current control and generates a torque proportional to the current, the actuator multiplies the torque value input as the control target value by a torque constant determined by the gear ratio. Torque control for generating input torque is realized by commanding the drive circuit. In particular, by providing a torque sensor at the active joint 123 and feeding back a value detected by the torque sensor to the drive circuit, highly accurate torque control is possible. Further, not only the rotary type but also a direct acting type actuator such as a hydraulic cylinder can be used.
  The detection mechanism 124 detects the state of the robot. The detection mechanism 124 is, for example, an encoder disposed at each joint, a floor reaction force sensor disposed at the foot, a gyro sensor disposed at the pelvis. The detection mechanism 124 may be an angle sensor that detects the angle of the joint, a posture sensor that acquires the posture of the robot, an external force sensor, or the like.
  The control unit 13 operates the active joint 123. The control unit 13 operates the active joint 123 in response to the torque or position command received by the receiving unit 11. For example, the control unit 13 converts the target floor acting force into each torque value of each actuator that drives each active joint 123 based on a forward kinematic model defined by the Jacobian matrix, and the like. Each converted torque value is output as a control target value to each actuator. The algorithm for determining the control target value that the control unit 13 outputs to the actuator is not limited. For example, it is described in Patent Document 1.
  It is preferable that the control unit 13 causes the electric motor 1232 to additionally operate with respect to a torque having a higher frequency than the first threshold value (assuming that “more” includes “more”). The additional operation is to operate the electric motor 1232 in addition to the air muscle 1231. Moreover, it is suitable for the control part 13 to carry out the additional operation of the electric motor 1232 with respect to the torque which is a load higher than a 2nd threshold value. Moreover, it is suitable for the control part 13 to perform additional operation of the electric motor 1232 with respect to the torque which is a higher frequency than a 1st threshold value and is a load higher than a 2nd threshold value. Here, the high frequency torque is, for example, a torque of a motion component. The high load torque is, for example, a torque greater than a gravity component (gravity compensation). That is, it is preferable that the control unit 13 uses only the air muscle 1231 for the gravity component (gravity compensation) torque and uses the air muscle 1231 and the electric motor 1232 for the motion component torque. . Here, the first threshold is, for example, 3 Hz. Further, the second threshold is, for example, 100 Nm.
The control unit 13 can usually be realized by an MPU, a memory, or the like. The processing procedure of the control unit 13 is usually realized by software, and the software is recorded on a recording medium such as a ROM. However, it may be realized by hardware (dedicated circuit).
The recording unit 131 records information such as programs and data required for control. The recording unit 131 holds, for example, the first threshold value and the second threshold value. The recording unit 131 can be realized by a recording medium such as a ROM, an EPROM, and a hard disk.
  The storage unit 132 temporarily stores data generated by executing the program. The storage unit 132 can be realized by a recording medium such as a RAM.
  The measuring means 133 accepts various signals (data) indicating detection results from the detection mechanism 124 such as a sensor.
  The control means 134 performs various calculations such as calculation of a control target value. The calculation performed by the control means 134 does not matter. The calculation method of the control means 134 will be described later.
  The output unit 135 outputs a signal to the active joint 123. For example, the output unit 135 outputs a target torque value to the active joint 123.
  Hereinafter, a prototype of the exoskeleton type robot 1 in the present embodiment will be described. The prototype of the exoskeleton robot 1 is a robot for walking / posture rehabilitation. A model diagram of the prototype of the exoskeleton robot 1 is shown in FIG.
The exoskeleton robot 1 has the following requirements (a) to (c) as hardware.
(A) Sufficient power and speed can be demonstrated (b) It can be introduced into a rehabilitation facility and operated in a wide space (c) The body is lightweight so that anyone can handle it easily
Further, the exoskeleton robot 1 has the following requirements (d) to (h) as software.
(D) Capable of accurate torque control (e) Having a safety function (f) Capable of basic autonomous movement (g) Capable of implementing various rehabilitation programs (h) Quickly determining the wearer's intention to exercise It can be extracted accurately
  As described above, the exoskeleton-type robot 1 positively employs a pneumatic and electric hybrid drive system as a drive system. Here, positive means the following three things. (1) Light weight; (2) Accurate torque control; (3) Optimal arrangement of air muscles considering the statics of the link mechanism.
  Here, with regard to the above (1), the air muscle is much lighter than the motor, air cylinder, hydraulic actuator, etc. of the same output, and a small brushless servo motor that has recently been remarkably improved in performance as an electric motor. Is available. In the prototype of the exoskeleton robot 1, the power source and the air compressor are installed outside, but the above requirement (b) is not particularly problematic.
  Regarding (2) above, it is essential to implement a rehabilitation program that is worn on the human body and has affinity for human motor control. In particular, life-size humanoid robots (M. Kawato, `` From `Understanding the Brain by Creating the Brain 'towards manipulative neuroscience,' 'Philosophical Transactions of the Royal Society, vol.363, no.1500, pp.2201-2214, 2008.) Full-body force control algorithm with proven track record (Xuan Zhao: "Optimal contact force control based on passivity of a humanoid robot with multiple grounding parts and redundant joints", Japanese robot Journal, vol. 27, no. 2, pp. 178-187, 2009. and Hyon, S., Morimoto, J. and Kawato, M., `` From compliant balancing to dynamic walking on humanoid robot: Integration of CNS and CPG '', IEEE ICRA, 2010 (in press).) Is considered effective. This algorithm assumes that joint torque can be accurately controlled. The exoskeleton-type robot 1 can increase the accuracy of torque control in a complex manner by having the air muscle take charge of the high load and low frequency torque and let the electric motor take charge of the low load and high frequency torque. For example, the former is assigned to the gravity component (statics) and the latter is assigned to the motion component (dynamics). In addition, since air muscle does not have stick-slip like an air cylinder, the movement is inherently very smooth.
  Regarding the above (3), the exoskeleton robot 1 is exposed to a high load to support 100% of the weight of both the wearer and the robot, but basically, gravity torque is almost unnecessary in an upright posture. Therefore, if the functional relationship between the displacement of the air muscle and the thrust is appropriately matched with the statics of the link mechanism, the weight of the entire mechanism can be reduced without sacrificing the output.
  FIG. 2 shows information indicating the configuration of the degree of freedom of the exoskeleton robot 1.
Here, the specifications of the prototype of the exoskeleton robot 1 are as follows.
(1) Including the base and lower body, it has a total of 10 active degrees of freedom including a total of 6 active degrees of freedom of the ankle, knee, and waist Pitch axes and 4 active degrees of freedom of the waist and ankle (see FIG. 2).
(2) All active joints are aerodynamic hybrid drive joints. The pneumatic hybrid drive joint is a joint having an air muscle 1231 and an electric motor 1232.
(3) It can be worn by an average Japanese person with a height of 170 cm, and the link length can be adjusted based on that.
(4) With a load assist rate of 0%, a normal person walks at normal speed and follows a full squat for 1 second in a round trip without resistance.
(5) With a load assist rate of 0%, a full squat for 2 seconds in a reciprocating manner is possible while maintaining dynamic balance.
  The exoskeleton-type robot 1 is equipped with an encoder at each joint, a floor reaction force sensor at the foot, and a gyro sensor at the pelvis as sensors. Moreover, it has a myoelectric sensor attached to a wearer and internal wiring. The control unit 13 (also referred to as a controller) reads values from the joint angle, floor reaction force, and gyro, and sets the joint torque target value as an air valve (valve for operating the air muscle 1231) of each axis (each active joint 123). The servo driver (driver that operates the electric motor 1232) is commanded. Then, the control unit 13 operates the air muscle 1231 and the electric motor 1232.
  Note that the active joint 123 is an aero-electric hybrid type as described above. That is, desirable torque controllability can be ensured by the air muscle 1231 and the electric motor 1232 of the active joint 123 cooperating according to the frequency and the magnitude of the load. Further, as shown in FIGS. 2, 3, etc., the air muscle 1231 is disposed at left and right ankles, left and right knees, and left and right positions of the waist. That is, the air muscle 1231 is optimally arranged. Therefore, high energy efficiency and simplification of control can be realized.
  Hereinafter, the effect of the static hybrid type active joint 123 will be described. FIG. 4 is a graph showing the characteristics of the hybrid drive realized by the active joint 123. In FIG. 4, the vertical axis represents torque and the horizontal axis represents time. The broken line (41) in FIG. 4 is a curve of torque generated by the electric motor 1232, the alternate long and short dash line (42) in FIG. 4 is a curve of torque generated by the air muscle 1231, and the solid line (43) in FIG. A solid line (43) in FIG. 4 indicates a torque (target torque) necessary for a certain motion. The solid line (43) in FIG. 4 indicates that the target torque changes dynamically during quick movement. Reference numeral 411 in FIG. 4 indicates an instantaneous rising torque that the electric motor 1232 is good at. Reference numeral 412 denotes air muscle steady-state error compensation by precise torque control by the electric motor 1232. Further, reference numeral 421 in FIG. 4 indicates a delay specific to the air muscle 1231. Further, the curve 42 in FIG. 4 indicates that the air muscle 1231 can continuously generate a small but large torque.
5 and 6 are schematic views of the hybrid drive active joint 123. FIG. 5 is a schematic diagram of a state when the joint air muscle 1231 has a natural length, and FIG. 6 is a schematic diagram of a state when the joint air muscle 1231 is in maximum contraction. 51 and 52 are robot links. 53 is a pulley. 54 is a wire. When air is introduced into the air muscle 1231 from the state of FIG. 5, the maximum contraction force is generated, and the link can be extended with a large torque. Moreover, even if air is put into the air muscle 1231 from the state of FIG. In the state shown in FIG. 6, no torque is necessary.
(Experimental result)
  In order to confirm the effectiveness of the hybrid drive method, we made a prototype of a test device simulating one foot, and performed a performance evaluation and a simple identification experiment.
  In the experiment, a simple link mechanism as shown in FIG. 7 was produced. In FIG. 7, the exoskeleton-type robot 1 includes a thigh, a lower leg, a bottom plate, and a movable top plate, and the top plate is smoothly restrained in the vertical direction by four-way guides. The weight is 12 kg for the top plate, 7 kg for the thigh, and 2 kg for the lower leg. The link length is 0.47 m. Here, the electric motor 1232 is disposed only at the knee joint. A DC servo motor with a gear head, which drives the knee via a belt pulley. A torque of about 20 Nm can be generated at a current of 5 A. On the other hand, the air muscle 1231 is attached to the knee and ankle. That is, only the knee is driven by aerodynamic hybrid. As an air muscle 1231 for the knee, Fest DSMP40-200 (basic length 200 mm) was adopted. This is a basic length and can exert a contraction force of 6000 N at a maximum pressure of 0.6 MPa. The tip of the air muscle 1231 is connected to the lower leg link via a wire and a pulley. Since the pulley radius is 404 mm, the torque of 240 Nm at maximum can be demonstrated with one air muscle 1231 in calculation. Since the actuator mass is at most 0.75 kg, the torque / weight ratio is overwhelmingly superior to other actuators. However, at the maximum contraction, the thrust becomes almost zero regardless of the pressure, and the torque cannot be exhibited. Therefore, it is important to provide the desired performance by optimally arranging the air muscle in consideration of the displacement-thrust characteristic and the static of the link mechanism. In this exoskeleton type robot 1, supporting the weight of the wearer is one of the most basic applications, but the gravitational torque is large when the robot is in a low posture and is small in the upright state. Therefore, by adjusting the end point of the wire so that the air muscle has the basic length in a low posture (the shape of the pulley is also changed if necessary), flat output characteristics can be provided over the entire range.
  The control unit 13 calculates a necessary pressure from the gravity compensation torque obtained by Equation 1 and the displacement of the air muscle, and performs pressure control using an electropneumatic proportional valve. In consideration of actual piping, the valve and the air muscle are separated by about 1 m. As a result of the experiment, it was confirmed that the top plate moves up and down smoothly by hand. That is, it was confirmed that the torque performance was good. Furthermore, using an ankle air muscle, it was confirmed that gravity compensation was possible even when a weight of 20 kg or more was placed.
As a task target in Formula 2, squat control was performed by giving a target height z (− is present immediately above z) as a COS function of 1 Hz. FIG. 8 shows the experimental results. Here, first, the control unit 13 performs control (gravity compensation and feedback control) in the work space only with the air muscle 1231, and turns on the electric motor 1232 in the second half. Here, the electric motor 1232 performs follow-up control on the z converted to the joint target trajectory q (− is present immediately above q). According to FIG. 8, after the electric motor 1232 is turned on, the trajectory followability is clearly improved (see 6 and thereafter on the horizontal axis of each graph in FIG. 8). This indicates that the electric motor 1232 compensates for the delay of the air muscle 1231. FIG. 9 shows a snapshot during the experiment. The air muscle is almost natural length at the lowest point and contracts by about 30% at the highest point.
  Next, an identification experiment of squat motion was performed. The air muscle 1231 has only gravity compensation, and the squat control is performed only by the electric motor 1232. This assumes that the motor torque is regarded as the torque generated by the wearer and the movement of the wearer is identified. However, unlike the simulation, the motion in the X direction is restricted, so there is only one degree of freedom, and the problem is trivial. Experimental data is shown in FIG. Gaussian noise with a standard deviation of 100 is superimposed on the input as an identification signal. As a result of applying a system identification algorithm called a subspace method, the system was identified as a single vibration system having poles of 0.9995 + 0.0315i and 0.9995-0.0315i. Of the dynamics of the robot, the known statics are compensated, so it can be said that the other dynamics are approximated.
  As described above, according to the present embodiment, by combining the pneumatic artificial muscle (air muscle 1231) and the electric motor 1232, it is possible to accurately control the high load torque required for the functional recovery and assist of walking and posture. A skeletal robot can be provided.
  In addition, according to the present embodiment, a torque having a necessary and sufficient size and accuracy can be exhibited. Therefore, even if a mannequin is mounted on a robot, posture control and walking can be freely performed by autonomous control.
  In addition, as shown in FIG. 10, the pneumatic hybrid drive of the active joint 123 employed in the present embodiment is very excellent as compared with other types of electric type, pneumatic type, and hydraulic type. In other words, the air-powered hybrid drive is a method that integrates the good points of the electric type and the pneumatic type.
  Moreover, according to this Embodiment, since the air muscle 1231 is optimally arrange | positioned, high energy efficiency and simplification of control are realizable.
  The present invention is not limited to the above-described embodiments, and various modifications are possible, and it goes without saying that these are also included in the scope of the present invention.
  As described above, the exoskeleton type robot according to the present invention has an effect that it can appropriately support exercise of the trunk and lower limbs, and is useful as an exoskeleton type robot for walking and posture rehabilitation. . Needless to say, if the structure does not allow people to enter, it can also be used as an autonomous humanoid robot. Because it is easy to increase the size of the actuator, it can be applied to heavy load robots that play an active role in fields such as factory transportation, building construction, environmental conservation, and disaster relief.
DESCRIPTION OF SYMBOLS 1 Exoskeleton type robot 11 Reception part 12 Exoskeleton 13 Control part 121 Base 122 Lower body 123 Active joint 124 Detection mechanism 131 Recording means 132 Storage means 133 Measurement means 134 Control means 135 Output means 1231 Air muscle 1232 Electric motor

Claims (4)

  1. An exoskeleton robot having a base and a lower body,
    Active joints that are active joints arranged at left and right ankles, left and right knees, and left and right positions of the waist,
    A control unit for operating the active joint,
    The active joint is
    An exoskeleton type robot equipped with an air muscle and an electric motor.
  2. The controller is
    Operating the air muscle and the electric motor with respect to a torque having a frequency higher than the first threshold;
    The exoskeleton robot according to claim 1, wherein only the air muscle is operated with respect to a torque having a frequency lower than the first threshold.
  3. The controller is
    Only the air muscle is operated for torque that is higher than the second threshold,
    The exoskeleton-type robot according to claim 1, wherein the air muscle and the electric motor are operated with respect to a torque having a lower load than the second threshold value.
  4. The controller is
    Only the air muscle is operated with respect to torque that is higher in load than the second threshold and lower in frequency than the first threshold,
    The exoskeleton-type robot according to claim 2 or 3, wherein the air muscle and the electric motor are operated with respect to a torque having a lower load than the second threshold and a higher frequency than the first threshold. .
JP2010190485A 2010-08-27 2010-08-27 Exoskeleton type robot Pending JP2012045194A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2010190485A JP2012045194A (en) 2010-08-27 2010-08-27 Exoskeleton type robot

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
JP2010190485A JP2012045194A (en) 2010-08-27 2010-08-27 Exoskeleton type robot

Publications (1)

Publication Number Publication Date
JP2012045194A true JP2012045194A (en) 2012-03-08

Family

ID=45900784

Family Applications (1)

Application Number Title Priority Date Filing Date
JP2010190485A Pending JP2012045194A (en) 2010-08-27 2010-08-27 Exoskeleton type robot

Country Status (1)

Country Link
JP (1) JP2012045194A (en)

Cited By (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102793619A (en) * 2012-07-04 2012-11-28 中国人民解放军海军航空工程学院 Statically determinate designing method for lower-limb exoskeleton suit structure
JP2014057628A (en) * 2012-09-14 2014-04-03 Advanced Telecommunication Research Institute International Power assist robot
JP2014057626A (en) * 2012-09-14 2014-04-03 Advanced Telecommunication Research Institute International Power assist robot
JP2014104549A (en) * 2012-11-28 2014-06-09 Advanced Telecommunication Research Institute International External skeleton robot, and rehabilitation device
JP2014155998A (en) * 2013-02-18 2014-08-28 Advanced Telecommunication Research Institute International Power assist robot
JP2014155653A (en) * 2013-02-18 2014-08-28 Advanced Telecommunication Research Institute International Power assist device and rehabilitation support device using the same
CN104302451A (en) * 2012-03-21 2015-01-21 雷沃克机器人技术有限公司 Motorized exoskeleton unit
CN104582668A (en) * 2012-06-15 2015-04-29 范德比尔特大学 Movement assistance device
WO2016039140A1 (en) * 2014-09-12 2016-03-17 株式会社国際電気通信基礎技術研究所 Actuator device, power assist robot, and humanoid robot
JP2016053824A (en) * 2014-09-03 2016-04-14 株式会社国際電気通信基礎技術研究所 Drive system
JP2016221680A (en) * 2016-09-23 2016-12-28 株式会社国際電気通信基礎技術研究所 Drive mechanism
CN106426097A (en) * 2016-10-26 2017-02-22 北京航空航天大学 Exoskeleton assisting mechanism for lower limb
EP3151399A4 (en) * 2014-05-27 2018-02-28 Advanced Telecommunications Research Institute International Actuator device, humanoid robot, and power assist device
CN109745159A (en) * 2019-01-21 2019-05-14 北京航空航天大学 A kind of ankle-joint artificial limb hydraulic drive circuit for realizing damping control and energy regenerating
US10315307B2 (en) 2015-07-09 2019-06-11 Hyundai Motor Company Wearable chair robot
WO2019134200A1 (en) * 2018-01-02 2019-07-11 北京理工大学 Flexible walking-aid exoskeleton
WO2019192636A1 (en) * 2018-04-06 2019-10-10 Juestel Peter Flight system

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007097636A (en) * 2005-09-30 2007-04-19 Matsushita Electric Ind Co Ltd Muscular strength assisting apparatus
JP2008087143A (en) * 2006-10-05 2008-04-17 Sony Corp Actuator control device

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2007097636A (en) * 2005-09-30 2007-04-19 Matsushita Electric Ind Co Ltd Muscular strength assisting apparatus
JP2008087143A (en) * 2006-10-05 2008-04-17 Sony Corp Actuator control device

Cited By (21)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104302451A (en) * 2012-03-21 2015-01-21 雷沃克机器人技术有限公司 Motorized exoskeleton unit
CN104582668A (en) * 2012-06-15 2015-04-29 范德比尔特大学 Movement assistance device
CN102793619A (en) * 2012-07-04 2012-11-28 中国人民解放军海军航空工程学院 Statically determinate designing method for lower-limb exoskeleton suit structure
JP2014057626A (en) * 2012-09-14 2014-04-03 Advanced Telecommunication Research Institute International Power assist robot
JP2014057628A (en) * 2012-09-14 2014-04-03 Advanced Telecommunication Research Institute International Power assist robot
JP2014104549A (en) * 2012-11-28 2014-06-09 Advanced Telecommunication Research Institute International External skeleton robot, and rehabilitation device
JP2014155998A (en) * 2013-02-18 2014-08-28 Advanced Telecommunication Research Institute International Power assist robot
JP2014155653A (en) * 2013-02-18 2014-08-28 Advanced Telecommunication Research Institute International Power assist device and rehabilitation support device using the same
EP3151399A4 (en) * 2014-05-27 2018-02-28 Advanced Telecommunications Research Institute International Actuator device, humanoid robot, and power assist device
EP3783781A1 (en) 2014-05-27 2021-02-24 Advanced Telecommunications Research Institute International Actuator device, humanoid robot and power assist device
JP2016053824A (en) * 2014-09-03 2016-04-14 株式会社国際電気通信基礎技術研究所 Drive system
CN106489030A (en) * 2014-09-12 2017-03-08 株式会社国际电气通信基础技术研究所 Actuator devices, power-assisted robot and anthropomorphic robot
WO2016039140A1 (en) * 2014-09-12 2016-03-17 株式会社国際電気通信基礎技術研究所 Actuator device, power assist robot, and humanoid robot
US10881536B2 (en) 2014-09-12 2021-01-05 Advanced Telecommunications Research Institute International Actuator device, power assist robot and humanoid robot
JP2016061302A (en) * 2014-09-12 2016-04-25 株式会社国際電気通信基礎技術研究所 Actuator device, power assist robot and humanoid robot
US10315307B2 (en) 2015-07-09 2019-06-11 Hyundai Motor Company Wearable chair robot
JP2016221680A (en) * 2016-09-23 2016-12-28 株式会社国際電気通信基礎技術研究所 Drive mechanism
CN106426097A (en) * 2016-10-26 2017-02-22 北京航空航天大学 Exoskeleton assisting mechanism for lower limb
WO2019134200A1 (en) * 2018-01-02 2019-07-11 北京理工大学 Flexible walking-aid exoskeleton
WO2019192636A1 (en) * 2018-04-06 2019-10-10 Juestel Peter Flight system
CN109745159A (en) * 2019-01-21 2019-05-14 北京航空航天大学 A kind of ankle-joint artificial limb hydraulic drive circuit for realizing damping control and energy regenerating

Similar Documents

Publication Publication Date Title
Jafari et al. A new actuator with adjustable stiffness based on a variable ratio lever mechanism
JP6688562B2 (en) Walking robot system that regenerates energy
US10449105B2 (en) System and method of bidirectional compliant joint torque actuation
Häufle et al. A clutched parallel elastic actuator concept: Towards energy efficient powered legs in prosthetics and robotics
KR101438970B1 (en) Method for controlling walking of robot
Hosoda et al. Pneumatic-driven jumping robot with anthropomorphic muscular skeleton structure
CN105105973B (en) Wearable power-assisted exoskeleton lower limb mechanism
Mao et al. Design of a cable-driven arm exoskeleton (CAREX) for neural rehabilitation
US9610208B2 (en) Device and method for decreasing energy consumption of a person by use of a lower extremity exoskeleton
Stienen et al. Self-aligning exoskeleton axes through decoupling of joint rotations and translations
Cestari et al. An adjustable compliant joint for lower-limb exoskeletons
JP6112567B2 (en) Exoskeleton
US9662262B2 (en) Joint torque augmentation system and method for gait assistance
US8070700B2 (en) Lower extremity enhancer
US8751041B2 (en) Method to generate humanlike motion of humanoid robot
Hosoda et al. Biped robot design powered by antagonistic pneumatic actuators for multi-modal locomotion
US8894592B2 (en) Device and method for decreasing oxygen consumption of a person during steady walking by use of a load-carrying exoskeleton
Kazerooni Exoskeletons for human power augmentation
Hobbelen et al. System overview of bipedal robots flame and tulip: Tailor-made for limit cycle walking
Zoss et al. Design of an electrically actuated lower extremity exoskeleton
CN103813771B (en) Exercise assist device and exercising support method, computer program and program recorded medium
Tsagarakis et al. A new variable stiffness actuator (CompAct-VSA): Design and modelling
Van Ham et al. Compliant actuator designs
Colbrunn et al. Design and control of a robotic leg with braided pneumatic actuators
Zoss et al. On the mechanical design of the Berkeley Lower Extremity Exoskeleton (BLEEX)

Legal Events

Date Code Title Description
A621 Written request for application examination

Free format text: JAPANESE INTERMEDIATE CODE: A621

Effective date: 20130821

A977 Report on retrieval

Free format text: JAPANESE INTERMEDIATE CODE: A971007

Effective date: 20140530

A131 Notification of reasons for refusal

Free format text: JAPANESE INTERMEDIATE CODE: A131

Effective date: 20140627

A02 Decision of refusal

Free format text: JAPANESE INTERMEDIATE CODE: A02

Effective date: 20141119