CN109946974B - Control system of electrically-driven quadruped robot - Google Patents
Control system of electrically-driven quadruped robot Download PDFInfo
- Publication number
- CN109946974B CN109946974B CN201910294462.1A CN201910294462A CN109946974B CN 109946974 B CN109946974 B CN 109946974B CN 201910294462 A CN201910294462 A CN 201910294462A CN 109946974 B CN109946974 B CN 109946974B
- Authority
- CN
- China
- Prior art keywords
- personal computer
- industrial personal
- joint
- driver
- encoder
- 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.)
- Active
Links
Images
Abstract
A control system of an electrically driven quadruped robot comprises an upper computer, an industrial personal computer, a driver and an encoder, wherein the industrial personal computer is connected with the upper computer to transmit data, and the industrial personal computer and the encoder are both connected with the driver; the encoder is fixed at the output end of the joint motor and used for detecting the rotating angle of the joint motor, the pulse number measured by the encoder and the reduction ratio of the joint speed reducer are used for obtaining the rotating angle of the joint, the pulse number is uploaded to the driver by the encoder, the filtered pulse value is uploaded to the industrial personal computer by the driver, then the industrial personal computer runs, the output torque of each joint of the robot is obtained and converted into a current value to be transmitted to the driver, the driver realizes the motor to output given torque through a built-in current ring, and the flexibility optimization and the collision energy optimization of a single leg are realized. The invention reduces the energy loss of single-leg grounding, improves the integral real-time performance and reliability of a hardware system and the high-speed motion of the quadruped robot, and can keep good dynamic stability and high energy efficiency.
Description
Technical Field
The invention relates to a control system for an electrically driven quadruped robot, and belongs to the technical field of control of the electrically driven quadruped robot.
Background
Leg-foot robots have been the more popular research field of mobile robots, and are superior to wheeled robots in both dynamic stability and complex terrain adaptability. The electrically driven robot has the above excellent characteristics as a hot branch of a legged robot. The electric drive robot is a robot which adopts a torque motor as a key unit for driving a joint to move, and is divided into a direct drive mode, a collimation drive mode and an SEA mode.
The electric drive robot in the direct drive mode directly connects the output shaft of the motor with the joint connecting rod, the drive mode has the best control effect, but the requirement on the torque motor is higher, the pure output torque of the existing torque motor cannot reach high torque performance, and the higher the performance is, the higher the cost is. The SEA mode is that a spring device is added between a torque motor and a connecting rod, and the high requirement on the performance of the torque motor is relieved by using the energy storage characteristic of a spring, but the dynamic model of the SEA mode is complex, and the actual control effect is difficult to achieve an ideal state. The quasi direct-drive mode is a scheme that a primary speed reducer is added between a motor and a connecting rod, so that the control effect of direct drive can be achieved, and the output torque of the motor can be amplified.
A two-degree-of-freedom quasi-direct-drive type single-leg structure of an electrically-driven quadruped robot (used in the present invention) is shown in fig. 1 and comprises a thigh joint 4, a thigh connecting rod 5, a shank joint 6 and a shank connecting rod 7. The structure of the whole simulation model of the electric-driven quadruped robot is shown in figure 2 and comprises a trunk 1 and four single legs with three degrees of freedom, wherein the trunk 1 is connected with a hip connecting rod 3 through a hip joint 2, the crotch connecting rod 3, a thigh joint 4, a thigh connecting rod 5, a shank joint 6 and a shank connecting rod 7 are sequentially connected, and a sole sensor 8 is arranged at the bottom of the shank connecting rod 7. The sole sensor 8 is used for detecting whether the foot end of one leg of the robot touches the ground or not. The thigh joint 4 and the shank joint 6 both consist of a torque motor and a speed reducer, the speed reducer is a primary planetary speed reducer, and the motor drives the thigh joint 4 or the shank joint 6 to rotate through the primary planetary speed reducer. Corresponding positive and inverse kinematics of the individual legs can be derived. Positive kinematics is to derive the position of the entire single-leg foot end in the coordinate system shown in fig. 1 by the motor angle transmitted by the encoder. The following is a specific analysis.
From the base coordinate system shown in fig. 1 and the parameters of the various joint links, the positive kinematic equation of the foot end relative to the base can be derived as follows:
knowing the position of the foot end of the single leg, the joint angle of the two-degree-of-freedom single leg can be obtained by inverse solution:
the derivation of the two-degree-of-freedom single leg can obtain a Jacobian matrix of the single leg:
the angle theta obtained in the above formulae1，θ2The joint angles of the thigh joint 4 and the calf joint 6, respectively, in fig. 1, x, z are the positions of the foot ends in the base coordinates. l1、l2The length of the thigh link 5 and the shank link 7, respectively. l. the12Is the linear distance from the foot end to the origin of the base coordinate system in figure 1.
In the field of quadruped robots, the establishment of a real-time control system is always a more central problem in the field of quadruped robots. Currently, there are many real-time system schemes, such as a real-time system built by RS-422 communication based on a controller provided by NI corporation and used by Cheetah2 of the national institute of technology, massachusetts, usa, a real-time system based on an ROS robot system and used by stareth of the national institute of technology, zurich, switzerland, and the like, or a mature scheme of the company or built based on the robot system. However, as no professional real-time system is adopted, the performance of the real-time system of the current electric leg-foot type robot is general, and the improvement aspect is still available.
In the field of robots, improving the energy utilization efficiency of robots is always a problem to be solved urgently. This problem is particularly relevant to electrically driven robots that focus primarily on energy optimization during foot end touchdown, foot end trajectory and selection of a quadruped gait.
Disclosure of Invention
The invention provides a control system of an electric drive quadruped robot, which is accurate, feasible and high in real-time performance, aiming at the problems of the existing electric drive quadruped robot in the aspects of real-time performance and high-efficiency control.
The control system of the electrically driven quadruped robot comprises thigh joints and shank joints, wherein the thigh joints and the shank joints are formed by connecting a motor and a speed reducer, and the control system adopts the following technical scheme:
the control system comprises an upper computer, an industrial personal computer, a driver and an encoder, wherein the industrial personal computer is connected with the upper computer to transmit data, and the industrial personal computer and the encoder are both connected with the driver; the encoder is fixed at the output end of the joint motor and used for detecting the rotating angle of the joint motor, the pulse number measured by the encoder and the reduction ratio of the joint speed reducer are used for obtaining the rotating angle of the joint, the pulse number is uploaded to the driver by the encoder, the filtered pulse value is uploaded to the industrial personal computer by the driver, then the industrial personal computer runs, the output torque of each joint of the robot is obtained and converted into a current value to be transmitted to the driver, the driver realizes the motor to output given torque through a built-in current ring, and the flexibility optimization and the collision energy optimization of a single leg are realized.
The upper computer realizes the interaction interface between the robot and the human. In the moving process of the robot, the upper computer can be used for giving control instructions to the robot, such as forward movement, jumping and the like.
The industrial personal computer runs a control algorithm to realize flexibility optimization and collision energy optimization of a single leg, and the control algorithm comprises single leg touchdown collision process optimization and four-foot running track and gait optimization; the single-leg touchdown collision is a continuous dynamics solving process, a parameter result is obtained by converting a continuous linear problem into a random linear problem and combining Gaussian distribution, and the optimized parameter expression is as follows:
the above equation is the basic Lagrangian kinetics equation, where M (q) is the Lagrangian quality matrix,
for mass acceleration, τ is output torque, FgAnd (3) calculating a final expression by combining Gaussian distribution for external force:
E[min(h，bG)2]＝h2-σ2(h+b)f(h)+(σ2+b2-h2)F(h)
wherein E represents the desired value, h, bGAre scalar parameters, σ is used to describe the mean distribution and degree of dispersion of the parameters of the process; f (h) is obtained by the following formula:
wherein f (t) is a desired function,
the function of the error is represented by,
the result of the energy loss of the touchdown collision is obtained;
the foot end trajectory is found by the application of a bernstein polynomial, as follows:
wherein P is1(t)，P2(t)，P3(t) x, z coordinates relative to the base coordinate system for a three-part piecewise program of the trajectory curve, t representing the time of the current system; a is0～a14X, z coordinates relative to a base coordinate system representing a predetermined point of the planned foot end curve;
the fly diagonal gait is derived by combining touchdown collision optimization and foot end trajectory curves, with two diagonal legs touching the ground and the other two diagonal legs emptying.
The industrial personal computer and the upper computer transmit data through a TCP/IP protocol, and the industrial personal computer is connected with the driver through a CAN bus. The encoder is connected with the driver through a signal line.
The calculation formula of the rotation angle theta of the robot joint is
Ni is the number of pulses detected by the encoder, and De is the reduction ratio of the robot joint reducer.
The control algorithm is mainly used for realizing the high-speed motion of the quadruped robot, and the simulation result shows that the control algorithm can enable the quadruped robot to reach the average speed of 18km/h, and can keep good dynamic stability and high energy efficiency.
Drawings
Fig. 1 is a schematic diagram for modeling the kinematics of a single-leg structure of a prior electrically-driven quadruped robot.
Fig. 2 is a schematic diagram of a simulation model of an electrically driven quadruped robot used in the prior art.
Fig. 3 is a schematic diagram of the structural principle of the control system in the present invention.
FIG. 4 is a schematic diagram of a trajectory planned by the control algorithm of the present invention.
Figure 5 is a schematic representation of a Flying-pitch gait.
In the figure: 1. the sensor comprises a trunk, 2 crotch joints, 3 crotch connecting rods, 4 thigh joints, 5 thigh connecting rods, 6 shank joints, 7 shank connecting rods and 8 sole sensors.
Detailed Description
The control system for the electrically driven quadruped robot aims at a two-degree-of-freedom single-leg structure shown in figure 1, and comprises an upper computer, an industrial personal computer, a driver and an encoder, wherein the industrial personal computer and the upper computer transmit data through a TCP/IP protocol, the industrial personal computer is connected with the driver through a CAN bus, the encoder is connected with the driver through a special signal line, and the encoder is fixed at the output end of a motor. The encoder is used for detecting the rotation angle of the motor, the pulse number measured by the encoder and the known reduction ratio De of the primary planetary reducer at the position of the lower leg joint in the figure 1 CAN be used for obtaining the rotation angle of the thigh joint and the lower leg joint of the legged robot in the figure 1, the encoder uploads the pulse number to the driver, the driver uploads the filtered pulse value to the industrial personal computer through the CAN communication bus in the figure 3 to realize the whole data acquisition process, then the industrial personal computer runs the control algorithm in the invention to obtain the output torque of each joint of the legged robot, the output torque is converted into a current value, the output torque is transmitted to the driver through CAN communication, and the driver realizes the output of given torque of the motor through a built-in current loop.
The industrial personal computer adopts a KEEX-5000 series industrial personal computer of German control and creation company and realizes CAN communication between the industrial personal computer and a driver through a CAN card of PEAK company extended by PCIE slots, a 12V direct current battery independently supplies power to the industrial personal computer and a cooling fan, and a 24V battery provided by Grignard is connected in series to reach 48V and is used as a single-leg direct current main power supply. The industrial computer is provided with a power indicator light and a CAN communication indicator light. As shown in fig. 3, the industrial personal computer is connected with the two drivers of the thigh joint 4 and the shank joint 6 of a single leg through a CAN bus, and data coding transmission and analysis are realized through a CANOpen protocol of a high-level protocol of the CAN. The system operated by the industrial personal computer is a QNX real-time system of blackberry company, and the QNX system has the characteristics of high real-time performance and reliability and is particularly suitable for the field of robots. The flexibility optimization and collision energy optimization of the single leg can be realized by compiling the control algorithm code in the invention through the specific QNX structural software.
The encoder adopts the relative formula encoder of reniersha model RMB20IC13BC10, and the precision is 8192, and every turn of motor promptly, 8192 pulses will be gathered to the encoder, and this type encoder precision is high, and is small, especially adapted motor angle's detection. The number of pulses detected by the encoder, using the formula:
the rotation angle theta of the joint can be obtained1，θ2Ni is the number of pulses detected by the encoder, and De is the reduction ratio of the first planetary reduction gear. The control algorithm in the present invention then calls on positive kinematics to calculate the position of the foot end relative to the base coordinate system and the instantaneous velocity from the known control frequency.
The system shows good real-time performance and high reliability performance in actual single-leg movement.
The control algorithm in the invention comprises optimization of the single-leg touchdown collision process and optimization of the four-foot running track and gait (jumping motion of single leg and Flying-trot gait motion of four-foot robot).
Single-leg touchdown collisions are a continuous dynamic solution process, but the input parameters available in real-world environments are insufficient to support an acceptable result. Therefore, a more accurate parameter result can be obtained by converting the continuous linear problem into the random linear problem and combining the Gaussian distribution. The optimized parameters are expressed as follows:
the above equation is a basic Lagrangian dynamics equation, where M (q) is the Lagrangian quality matrix,
for mass acceleration, τ is output torque, FgIn order to combine external force and Gaussian distribution, the final expression can be obtained:
E[min(h，bG)2]＝h2-σ2(h+b)f(h)+(σ2+b2-h2)F(h)
wherein E represents the desired value, h, bGAre scalar parameters and σ is used to describe the mean distribution and degree of dispersion of the parameters of the process. F (h) is obtained by the following formula:
wherein f (t) is a desired function,
an error function is represented. This results in a lower energy loss for touchdown collisions.
Referring to fig. 4, a better foot end trajectory can be found by applying bernstein polynomial, and the trajectory expression is as follows:
wherein P is1(t)，P2(t)，P3(t) x, z coordinates relative to the base coordinate system for a three-part piecewise program of the trajectory plot, t representing the time of the current system. a is0～a14Representing the x, z coordinates of the pre-set points of the planned foot end curve relative to the base coordinate system.
As shown in fig. 5, gait results in a "fly diagonal" gait by combining touchdown collision optimization with trajectory profiling, i.e., two diagonal legs are touchdown, while the other two diagonal legs are vacated. The black circles in fig. 5 represent the corresponding three-degree-of-freedom single-leg touchdown and the white circles represent the corresponding three-degree-of-freedom single-leg loft. The overall diagram shows the single leg touchdown variation of a one cycle quadruped robot.
The above are all of the specific control algorithms. The algorithm mainly realizes the high-speed motion of the quadruped robot, and the simulation result shows that the control algorithm of the invention can enable the quadruped robot to reach the average speed of 18km/h and can maintain good dynamic stability and high energy efficiency.
Claims (1)
1. A control system of an electrically driven quadruped robot is characterized in that: the industrial personal computer is connected with the upper computer to transmit data, and the industrial personal computer and the encoder are both connected with the driver; the encoder is fixed at the output end of the joint motor and used for detecting the rotation angle of the joint motor, the rotation angle of the joint is obtained through the pulse number measured by the encoder and the reduction ratio of the joint reducer, the pulse number is uploaded to the driver by the encoder, the filtered pulse value is uploaded to the industrial personal computer by the driver, then the industrial personal computer runs to obtain the output torque of each joint of the robot and convert the output torque into a current value to be transmitted to the driver, the driver realizes the motor to output a given torque through a built-in current ring, and the flexibility optimization and the collision energy optimization of a single leg are realized;
the industrial personal computer is connected with the driver through a CAN bus, the industrial personal computer and the upper computer transmit data through a TCP/IP protocol, the industrial personal computer and the cooling fan are separately powered by a 12V direct current battery, and a 24V battery is connected in series to achieve 48V and is used as a single-leg direct current main power supply; the industrial personal computer is connected with the drivers through a CAN bus, is connected with the two drivers of the thigh joint and the shank joint of the single leg through the CAN bus, and realizes data coding transmission and analysis through a high-level protocol CANOpen protocol of the CAN;
the pulse number measured by the encoder is converted into the position of the motor, the rotation angle of the joint is obtained through the reduction ratio of the joint reducer, the output value of the filtered encoder is uploaded to the industrial personal computer by the driver, then the industrial personal computer runs to obtain the output torque of each joint of the robot and convert the output torque into a current value to be transmitted to the driver, the driver realizes that the motor outputs given torque through a built-in current loop, and the thigh motor and the shank motor are converted into joint output torque through the planetary reducer and the rigid connecting rod;
the system operated by the industrial personal computer is a QNX real-time system, high real-time performance of control is guaranteed, the effect of energy consumption optimization is guaranteed through high real-time control, and the control algorithm of the industrial personal computer is compiled through specific QNX structural software to realize active flexible force control on a single leg, so that flexible optimization and collision energy optimization of a single-leg movement flying gait are realized;
the industrial personal computer runs a control algorithm to realize flexibility optimization and collision energy optimization of a single leg, and the control algorithm comprises single leg touchdown collision process optimization and four-foot running track and gait optimization; the single-leg touchdown collision is a continuous dynamics solving process, a parameter result is obtained by converting a continuous linear problem into a random linear problem and combining Gaussian distribution, and the optimized parameter expression is as follows:
the above equation is the basic Lagrangian kinetics equation, where M (q) is the Lagrangian quality matrix,
for mass acceleration, τ is output torque, FgAnd (3) calculating a final expression by combining Gaussian distribution for external force:
E[min(h，bG)2]＝h2-σ2(h+b)f(h)+(σ2+b2-h2)F(h)
wherein E represents the expected value, h, bGAre scalar parameters, σ is used to describe the mean distribution and degree of dispersion of the parameters of the process; f (h) is obtained by the following formula:
wherein f (t) is a desired function,
the function of the error is represented by,
the result of the energy loss of the touchdown collision is obtained;
the foot end trajectory is found by the application of a bernstein polynomial, as follows:
wherein P is1(t)，P2(t)，P3(t) x, z coordinates relative to the base coordinate system for a three-part piecewise program of the trajectory curve, t representing the time of the current system; a is0～a14X, z coordinates relative to a base coordinate system representing a predetermined point of the planned foot end curve;
the fly diagonal gait is derived by combining touchdown collision optimization and foot end trajectory curves, with two diagonal legs touching the ground and the other two diagonal legs emptying.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910294462.1A CN109946974B (en) | 2019-04-12 | 2019-04-12 | Control system of electrically-driven quadruped robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910294462.1A CN109946974B (en) | 2019-04-12 | 2019-04-12 | Control system of electrically-driven quadruped robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109946974A CN109946974A (en) | 2019-06-28 |
| CN109946974B true CN109946974B (en) | 2022-07-01 |
Family
ID=67014915
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910294462.1A Active CN109946974B (en) | 2019-04-12 | 2019-04-12 | Control system of electrically-driven quadruped robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109946974B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110412921B (en) * | 2019-08-09 | 2021-07-27 | 山东大学 | Robot single-leg high-real-time control system based on EtherCAT |
| CN111332382A (en) * | 2020-04-01 | 2020-06-26 | 安徽工业大学 | Horizontal joint quadruped robot |
| CN111890363B (en) * | 2020-07-27 | 2022-12-30 | 四川大学 | Mechanical arm motion planning method based on rapid self-adaptive gradient neural network algorithm |
| CN111857173A (en) * | 2020-08-17 | 2020-10-30 | 常州工程职业技术学院 | Jumping gait planning control system and method for quadruped robot |
| CN113110304A (en) * | 2021-03-13 | 2021-07-13 | 无锡信捷电气股份有限公司 | Multi-channel sport |
| CN113377113B (en) * | 2021-07-02 | 2022-01-04 | 北方工业大学 | Foot end trajectory planning method and system of foot type robot and control method and system |
| CN114488808B (en) * | 2022-01-24 | 2022-12-30 | 新基线(江苏)科技有限公司 | Real-time correction method for model parameters of four-footed robot |
| CN114454983B (en) * | 2022-03-02 | 2023-04-14 | 北京理工大学 | Turning control method and system for quadruped robot |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4548135B2 (en) * | 2005-02-03 | 2010-09-22 | トヨタ自動車株式会社 | Legged robot and its control method |
| CN103676797A (en) * | 2012-09-07 | 2014-03-26 | 南京理工大学 | Modularized sub-action multi-legged robot motion controller and control method thereof |
| CN105599821A (en) * | 2016-01-06 | 2016-05-25 | 山东优宝特智能机器人有限公司 | Electric driven biomimetic four-leg robot with environment sensing ability and control method |
| CN108710298A (en) * | 2018-05-23 | 2018-10-26 | 中国海洋大学 | Low energy consumption posture analysis method based on sufficient formula bio-robot under quiet vertical state |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8457830B2 (en) * | 2010-03-22 | 2013-06-04 | John R. Goulding | In-line legged robot vehicle and method for operating |
-
2019
- 2019-04-12 CN CN201910294462.1A patent/CN109946974B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4548135B2 (en) * | 2005-02-03 | 2010-09-22 | トヨタ自動車株式会社 | Legged robot and its control method |
| CN103676797A (en) * | 2012-09-07 | 2014-03-26 | 南京理工大学 | Modularized sub-action multi-legged robot motion controller and control method thereof |
| CN105599821A (en) * | 2016-01-06 | 2016-05-25 | 山东优宝特智能机器人有限公司 | Electric driven biomimetic four-leg robot with environment sensing ability and control method |
| CN108710298A (en) * | 2018-05-23 | 2018-10-26 | 中国海洋大学 | Low energy consumption posture analysis method based on sufficient formula bio-robot under quiet vertical state |
Non-Patent Citations (5)
| Title |
|---|
| An Energy Efficient Motion Controller Based on SLCP for the Electrically Actuated Quadruped Robot;Tianfa Li etc.;《Journal of Bionic Engineering》;20200407;全文 * |
| Stochastic Complementarity for Local Control of Dscontinuous Dynamics;Yuval Tassa etc.;《Robotics: Science and Systems VI》;20100627;全文 * |
| 主—被动变刚度柔性单腿建模及能耗分析;史延雷 等;《计算机集成制造系统》;20171231;第23卷(第12期);第2719-2726页 * |
| 史延雷 等.主—被动变刚度柔性单腿建模及能耗分析.《计算机集成制造系统》.2017,第23卷(第12期),第2719-2726页. * |
| 基于自由步态的六足机器人崎岖路面行走控制方法研究;丛恩博;《中国优秀硕士学位论文全文数据库 信息科技辑》;20180215(第2期);正文第33-34页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109946974A (en) | 2019-06-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109946974B (en) | Control system of electrically-driven quadruped robot | |
| Park et al. | Variable-speed quadrupedal bounding using impulse planning: Untethered high-speed 3d running of mit cheetah 2 | |
| CN101916071B (en) | CPG feedback control method of biomimetic robot fish movement | |
| Chen et al. | Quattroped: a leg--wheel transformable robot | |
| CN110239644A (en) | A kind of bionical quadruped robot based on flexible spinal technology | |
| CN109032138A (en) | Multi-robot formation control system and method based on consistency algorithm | |
| CN104108433A (en) | Compliant control method of wheel foot type robot | |
| CN111746824B (en) | Buffer/walking integrated hexapod lander and gait control method thereof | |
| CN106809294A (en) | The walking robot that cam link combined mechanism drives | |
| CN113127989A (en) | Six-degree-of-freedom mechanical arm inverse kinematics analysis solution control method | |
| CN110181480A (en) | A kind of multi-joint is helped the elderly invalid wheelchair mechanical arm and its Kinematic Model method based on spinor theory | |
| CN103412487A (en) | Spherical robot slope motion control method based on fuzzy sliding mode controller | |
| CN110919656A (en) | SLIP-based parallel leg and foot robot motion control method and system | |
| Liao et al. | Energy efficient swing leg trajectory planning for quadruped robots walking on rough terrain | |
| CN111497964B (en) | Distributed control system of electrically-driven quadruped robot | |
| CN102541068A (en) | Lower limb motion planning system for biped robot in obstacle crossing | |
| Sun et al. | Rhex-t3: A transformable hexapod robot with ladder climbing function | |
| Li et al. | Design and simulated characteristics of a new biped mechanism | |
| CN102126220A (en) | Control system for six-degree-of-freedom mechanical arm of humanoid robot based on field bus | |
| CN204819543U (en) | Centralized control formula multirobot motion control system | |
| CN107485540B (en) | Energy injection system for intelligent walking stick | |
| CN203899136U (en) | Exterior structure of embedded vision-based competitive entertainment-type humanoid robot | |
| CN206494025U (en) | Imitative spider robot | |
| CN108673492A (en) | Modularization soft robot system based on webpage control | |
| CN209793754U (en) | Direct-drive human-simulated finger convenient to control |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |