ROS-based simulation control platform and exoskeleton robot simulation control system
Through the ROS-based simulation control platform, the problem of insufficient motion intention recognition and control performance of exoskeleton robots in complex environments is solved, and the rapid verification and parameter debugging of exoskeleton robot control algorithms are realized, improving operational efficiency and safety.
Patent Information
- Application Number
- CN202110792420.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-07-14
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2041-07-14
AI Technical Summary
In the prior art, exoskeleton robots are difficult to effectively identify the wearer's movement intention and improve control performance under the complexity of ring and operating environment, resulting in insufficient operating efficiency and safety.
Adopting a simulation control platform based on ROS, including control module, data acquisition module, simulation display module and motion intention detection module, data acquisition module is collected in real time through PCI-CAN board, building an exoskeleton robot model, supporting customized dynamic mathematical models and open source A three-dimensional physics simulation platform, performs simulation display and algorithm verification, and realizes online controller parameter debugging.
The development cycle of exoskeleton software control algorithm is shortened, experimental verification of exoskeleton robot control algorithm and debugging of controller parameters is realized, and control performance and safety are improved.
Smart Images

Figure CN113608451B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of exoskeleton robots, and in particular to a simulation control platform based on ROS and an exoskeleton robot simulation control system. Background Art
[0002] The power-assisted lower limb exoskeleton robot is a wearable human-machine integrated mechanical device that can provide power assistance through electric and other means to enhance the power output of the human body. It is suitable for assisting heavy manual laborers or helping the elderly and the disabled. At the same time, the exoskeleton robot can also be used in special fields such as single-soldier combat and emergency rescue. It is used to assist rescuers or soldiers to carry related equipment for long distances in scenes where conventional vehicles cannot enter. Therefore, exoskeleton robot technology has important significance and value in the future fields of intelligent human-machine interaction and human-machine integration.
[0003] The motion control of lower limb exoskeleton robots is the premise and guarantee for them to assist the wearer in carrying and transporting tasks, and is one of the key technologies of exoskeleton robots. Due to the complexity of the wearer's environment and the operating environment of the exoskeleton robot, identifying the wearer's motion intention and improving its control performance have become the key to conquering exoskeleton robot technology. However, during the development and operation of exoskeleton robots, due to the presence of operators, conventional offline programming and debugging methods are difficult to guarantee efficiency and safety. Therefore, it is necessary to propose further solutions to the above problems. Summary of the invention
[0004] The purpose of the present invention is to provide a simulation control platform and an exoskeleton robot simulation control system based on ROS to overcome the deficiencies in the prior art.
[0005] To achieve the above-mentioned object of the invention, the present invention provides a ROS-based simulation control platform for simulation control of an exoskeleton robot, the ROS-based simulation control platform comprising:
[0006] Control module;
[0007] A data acquisition module, which collects data fed back by the exoskeleton robot and transmits the data to the control module;
[0008] A simulation display module, which constructs an exoskeleton robot model and performs simulation display of the exoskeleton robot model under the control of the control system based on the feedback data;
[0009] The motion intention detection module includes: a plurality of motion intention detection units, which subscribe to the data fed back by the exoskeleton robot and perform human motion intention recognition.
[0010] As an improvement of the ROS-based simulation control platform of the present invention, the data acquisition module collects data from the exoskeleton robot in real time with the help of a PCI-CAN board.
[0011] As an improvement of the ROS-based simulation control platform of the present invention, the data acquisition module publishes the collected data in the form of a ROS topic, and at the same time subscribes to the joint control information of the control module and sends it to the exoskeleton robot.
[0012] As an improvement of the ROS-based simulation control platform of the present invention, the control module supports a custom exoskeleton dynamics mathematical model or an open source three-dimensional physical simulation platform.
[0013] As an improvement of the ROS-based simulation control platform of the present invention, when the control module supports the open source three-dimensional physical simulation platform, the simulation control process is executed according to the following steps:
[0014] The exoskeleton three-dimensional structure diagram is imported into the open source three-dimensional physical simulation platform, and then the information parameters of the constructed exoskeleton robot model are adjusted and optimized. At the same time, with the help of the control algorithm nodes of the open source three-dimensional physical simulation platform, the model and the actual system in the platform are controlled and simulated using a low-order control algorithm;
[0015] Verify the relevant torque depth control algorithm and observe the simulation operation effect in the simulation platform. After completing the control algorithm verification, use the control information sent to the lower exoskeleton end to perform online algorithm verification tests on the actual system end.
[0016] After the algorithm is verified, the exoskeleton dynamics optimization model is built based on the actual system's computing power configuration, and compared and verified on the simulation platform.
[0017] As an improvement of the ROS-based simulation control platform of the present invention, the PC simulation control platform decomposes the movement state into: left foot vacant and right foot supporting state, double-foot supporting state and right foot vacant and left foot supporting state during the exoskeleton human-machine coupled movement.
[0018] As an improvement of the ROS-based simulation control platform of the present invention,
[0019] Single-leg support dynamics model:
[0020]
[0021] Where:
[0022] is the sum of the kinetic energy of each member of the exoskeleton;
[0023] is the sum of the gravitational potential energy of each member of the exoskeleton;
[0024] is the elastic potential energy of the exoskeleton knee joint and;
[0025] L is the Lagrangian function, which is equal to the difference between the total kinetic energy of the system and the gravitational potential energy and the total elastic potential energy;
[0026] Respectively represent the generalized angular velocity of the knee and ankle joints;
[0027] The generalized coordinates that characterize the sagittal and frontal plane systems, respectively;
[0028] They respectively represent the combined force of the knee and ankle joints, namely the joint motor driving force and the human-computer interaction force.
[0029] As an improvement of the ROS-based simulation control platform of the present invention,
[0030] The 6-link dynamic equation of the exoskeleton support phase is:
[0031]
[0032] In the formula, represents the inertia tensor matrix of the robot in the sagittal plane;
[0033] is a 6x6 matrix of Coriolis and centrifugal forces;
[0034] is a 6 x1 gravity matrix;
[0035] is the elastic force matrix;
[0036] , , They represent the 6x1-dimensional generalized position, generalized velocity, and generalized acceleration respectively.
[0037] As an improvement of the ROS-based simulation control platform of the present invention, in the double-foot support state, the output torque of the left and right hip joints can be calculated based on the distance between the projection point of the exoskeleton's overall center of gravity on the line connecting the two feet's landing points and the two feet. , Proportional distribution:
[0038] , is the total torque required by the left and right hip joints of the exoskeleton system;
[0039] The other joint torques are derived from the 5-link dynamics:
[0040]
[0041] In the formula, is the kinetic energy of each link of the exoskeleton, including the left and right calves, left and right thighs, and torso;
[0042] is the potential energy sum of each link of the exoskeleton, including the left and right calves, left and right thighs, and torso;
[0043] Es is the elastic potential energy of the knee joint of the system.
[0044] To achieve the above-mentioned purpose of the invention, the present invention provides a lower limb exoskeleton robot simulation control system based on ROS, which comprises:
[0045] The lower exoskeleton side, the ROS-based simulation control platform as described above;
[0046] The lower exoskeleton end includes: a human posture sensing module, a joint drive motor module and an interactive force detection module, and the ROS-based simulation control platform can receive data fed back from the human posture sensing module and the interactive force detection module;
[0047] The human posture sensing module includes: a six-axis inertial measurement sensor arranged on a back frame, an angle sensor arranged at the knee joints of the left and right legs, and a pressure sensor arranged at the soles of the left and right feet.
[0048] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention provides a ROS-based simulation control platform and an exoskeleton robot simulation control system, which can shorten the development cycle of the exoskeleton software control algorithm, and can realize experimental verification of the exoskeleton robot control algorithm by simulating a real exoskeleton robot model, and can also provide debugging of controller parameters. BRIEF DESCRIPTION OF THE DRAWINGS
[0049] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0050] Figure 1 It is a module diagram of an embodiment of an exoskeleton robot simulation control system based on ROS of the present invention;
[0051] Figure 2 This is the simulation control flow chart when the control module supports the open source 3D physical simulation platform. DETAILED DESCRIPTION
[0052] The present invention is described in detail below in conjunction with various embodiments, but it should be noted that these embodiments are not limitations of the present invention, and any equivalent transformations or substitutions in functions, methods, or structures made by ordinary technicians in the field based on these embodiments are all within the scope of protection of the present invention.
[0053] like Figure 1 As shown, an embodiment of the present invention provides an exoskeleton robot simulation control system based on ROS, which adopts a modular distributed master-slave control hardware architecture. The exoskeleton robot simulation control system includes: an upper-layer ROS-based simulation control platform 100 and a lower-layer exoskeleton terminal 200. The upper-layer ROS-based simulation control platform 100 performs simulation control and task decision-making, and the lower-layer exoskeleton terminal 200 performs control according to the instructions of the control platform. The overall system presents the characteristics of high cohesion and low coupling.
[0054] Specifically, the PC simulation control platform is built based on the ROS environment running on the real-time Linux operating system, which includes: a control module 101, a data acquisition module 102, a simulation display module 103 and a motion intention detection module 104.
[0055] The control module 101 may be a PC host, which is mainly responsible for sensor data collection, motion simulation and control of the exoskeleton robot system. At the same time, to ensure real-time performance, the PC host runs the ROS2 system, which uses a communication framework based on the Data Distribution Service (DDS).
[0056] The data acquisition module 102 collects the data fed back by the lower exoskeleton end 200 and transmits the data to the control module 101. The data acquisition module 102 collects the data of the exoskeleton robot end in real time with the help of the PCI-CAN board. At the same time, the data acquisition module 102 publishes the collected data in the form of ROS topics, subscribes to the joint control information of the control module 101, and sends it to the lower exoskeleton end 200.
[0057] The simulation display module 103 is used to construct an exoskeleton robot model, and based on the feedback data, simulate and display the exoskeleton robot model under the control of the control system. Specifically, the simulation display module 103 can write a URDF file based on the actual exoskeleton robot physical model, construct an exoskeleton robot model, and then import the exoskeleton sensor information into the robot simulation environment RVIZ in the ROS system, and can directly display the motion simulation screen of the current exoskeleton robot during the exoskeleton debugging operation.
[0058] The motion intention detection module 104 is used for detecting the motion state of a human body and identifying the motion intention. The motion intention detection module 104 comprises: a plurality of motion intention detection units, which subscribe to the data fed back by the exoskeleton robot and perform motion intention identification of a human body.
[0059] Furthermore, the control module 101 supports a custom exoskeleton dynamics mathematical model or an open source three-dimensional physical simulation platform (Gazebo). Gazebo is equipped with an ODE physics engine that simulates real physics. ODE is an industrial-quality rigid body dynamics library that can create a virtual simulation environment for the exoskeleton robot. In addition, Gazebo supports sensor simulation and has a wealth of expandable plug-ins.
[0060] like Figure 2 As shown, when the control module 101 supports the open source three-dimensional physical simulation platform, the simulation control process is executed according to the following steps:
[0061] S1. Directly import the exoskeleton 3D structure diagram into the Gazebo simulation platform through the SW2URDF plug-in, and then adjust and optimize the information parameters such as the quality of the exoskeleton model based on the actual exoskeleton system. At the same time, the control algorithm node can be opened, and the low-order control algorithm can be used to control and simulate the model and the actual system in the platform, and the model parameters can be identified and corrected based on the operation effect, so that the rigid body dynamics model in the simulation platform is closer to the actual exoskeleton system.
[0062] S2. You can directly use Gazebo's ros_control to conduct verification research on the relevant torque depth control algorithm, and observe the simulation operation effect in the simulation platform. After completing the control algorithm verification, the user can enable the control information sent to the lower exoskeleton end 200 and perform online algorithm verification tests on the actual system end.
[0063] S3. After the algorithm is verified, you can build an exoskeleton dynamics optimization model based on the actual system's computing power configuration to replace the Gazebo node to reduce the system load requirements, and perform comparative verification on the simulation platform.
[0064] In the process of human-machine coupling movement of the exoskeleton, based on the rhythmicity of human walking, the movement state can be decomposed into three basic movement configurations: the left foot is free and the right foot is supporting the state, the double-foot supporting state, and the right foot is free and the left foot is supporting the state. The human-machine coupling movement process is the switching process of these three typical movement configurations. With the switching of movement and movement configuration, the dynamic structure of the system will change. Therefore, the dynamic equations of the exoskeleton robot need to be established according to different movement configurations.
[0065] For the single-leg support dynamics model:
[0066]
[0067] Where:
[0068] is the sum of the kinetic energy of each member of the exoskeleton;
[0069] is the sum of the gravitational potential energy of each member of the exoskeleton;
[0070] is the elastic potential energy of the exoskeleton knee joint and;
[0071] L is the Lagrangian function, which is equal to the difference between the total kinetic energy of the system and the gravitational potential energy and the total elastic potential energy;
[0072] Respectively represent the generalized angular velocity of the knee and ankle joints;
[0073] The generalized coordinates that characterize the sagittal and frontal plane systems, respectively;
[0074] Respectively represent the combined force of the knee and ankle joints, namely the joint motor driving force and the human-machine interaction force;
[0075] In the single-leg support phase, the system can be viewed as a 6-link model, including left and right calf rods, left and right thigh rods, torso and supporting foot, and the 4 active degrees of freedom are located at the left and right hip and knee joints.
[0076] Therefore, the 6-link dynamic equation of the exoskeleton support phase is:
[0077]
[0078] In the formula, represents the inertia tensor matrix of the robot in the sagittal plane;
[0079] is a 6x6 matrix of Coriolis and centrifugal forces;
[0080] is a 6 x1 gravity matrix;
[0081] is the elastic force matrix;
[0082] , , They represent the 6x1-dimensional generalized position, generalized velocity and generalized acceleration respectively;
[0083] The force exerted by the human body on the exoskeleton is , obtained by the human-machine interaction force sensor of the system, which is a 6x1 matrix;
[0084] J is a 6x6 matrix, representing the mapping of the Cartesian space force on the sagittal plane of the exoskeleton to the robot joint space.
[0085] For double-foot support:
[0086] The output torque of the left and right hip joints can be calculated based on the distance between the projection point of the exoskeleton's overall center of gravity on the line connecting the two feet's landing points and the two feet. , Proportional distribution:
[0087] , is the total torque required by the left and right hip joints of the exoskeleton system;
[0088] The other joint torques are derived from the 5-link dynamics:
[0089]
[0090] In the formula, is the kinetic energy of each link of the exoskeleton, including the left and right calves, left and right thighs, and torso;
[0091] is the potential energy sum of each link of the exoskeleton, including the left and right calves, left and right thighs, and torso;
[0092] Es is the elastic potential energy of the knee joint of the system.
[0093] The lower exoskeleton end 200 includes an exoskeleton hip and knee joint motor module control system. The drivers of each joint motor are integrated in the joint control node board, and the position, speed and torque mode operation settings of the joint motor can be realized at the bottom layer. The lower exoskeleton end 200 and the upper ROS-based simulation control platform 100 are connected by CAN bus. The protocol layer uses CANopen protocol, which supports flexible addition and subtraction of hip and knee active joint motor nodes.
[0094] Specifically, the lower exoskeleton end 200 includes: a human posture sensing module 201 , a joint driving motor module 202 and an interaction force detection module 203 .
[0095] The ROS-based simulation control platform can receive data feedback from the human posture sensing module 201 and the interactive force detection module 203. The human posture sensing module 201 includes: a six-axis inertial measurement sensor arranged on the back frame, an angle sensor arranged at the knee joints of the left and right legs, and a pressure sensor arranged at the left and right soles of the feet. Each sensor is fed back to the several motion intention detection units in the above-mentioned motion intention detection module 104 as the data basis for human motion state detection and motion intention recognition. The joint drive motor module 202 can be set to multiple groups as needed.
[0096] In one implementation, an IMU six-axis inertial measurement sensor is mounted on the back to detect the spatial posture information of the upper torso; an absolute encoder is built into each knee and ankle joint of the left and right legs, a total of 6 encoders, which can be used to detect the angle information of the knee and ankle joints; at the same time, a plantar pressure sensor is installed on the left and right soles of the exoskeleton, a total of 2 sensors, which can be used to detect the state of lifting and lowering the legs. Based on this, developers can collect sensor information from the exoskeleton end online to analyze the state of standing, sitting, walking, bending, etc. of the wearable exoskeleton system.
[0097] In summary, the present invention provides a simulation control platform and an exoskeleton robot simulation control system based on ROS. The system can shorten the development cycle of the exoskeleton software control algorithm, and can realize the experimental verification of the exoskeleton robot control algorithm by simulating the real exoskeleton robot model. At the same time, it can also provide debugging of controller parameters.
[0098] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the present invention can be implemented in other specific forms without departing from the spirit or essential features of the present invention. Therefore, the embodiments should be considered exemplary and non-restrictive in all respects, and the scope of the present invention is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present invention.
[0099] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.
Claims
1. A simulation control platform based on ROS, which is used for the simulation control system of an exoskeleton robot, characterized in that: The ROS-based simulation control platform includes: Control module; A data acquisition module, which collects data fed back by the exoskeleton robot and transmits the data to the control module; A simulation display module, which constructs an exoskeleton robot model and performs simulation display on the exoskeleton robot model under the control of the simulation control system based on the feedback data; A motion intention detection module, comprising: a plurality of motion intention detection units, wherein the plurality of motion intention detection units subscribe to the data fed back by the exoskeleton robot and perform human motion intention recognition; The control module supports a custom exoskeleton dynamics mathematical model or an open source three-dimensional physical simulation platform; when the control module supports the open source three-dimensional physical simulation platform, the simulation control process is executed according to the following steps: S1. Directly import the exoskeleton 3D structure diagram into the Gazebo simulation platform through the SW2URDF plug-in, and then adjust and optimize the exoskeleton model quality information parameters based on the actual exoskeleton system; at the same time, the control algorithm node can be opened, and the low-order control algorithm can be used to control and simulate the model in the platform and the actual exoskeleton system, and the model parameters can be identified and corrected based on the operation effect, so that the rigid body dynamics model in the simulation platform is closer to the actual exoskeleton system; S2. Use Gazebo's ros_control to verify the relevant torque depth control algorithm and observe the simulation running effect in the simulation platform. After completing the control algorithm verification, the user enables the control information sent to the lower-level exoskeleton end and performs online algorithm verification tests on the actual exoskeleton system end. S3. After the algorithm is verified, based on the actual computing power configuration of the exoskeleton system, an exoskeleton dynamics optimization model is built to replace the Gazebo node to reduce the load requirements of the actual exoskeleton system, and a comparative verification is performed on the simulation platform.
2. The ROS-based simulation control platform according to claim 1, characterized in that: The data acquisition module uses the PCI-CAN board to collect data from the exoskeleton robot in real time.
3. The ROS-based simulation control platform according to claim 2, characterized in that: The data acquisition module publishes the collected data in the form of a ROS topic, and at the same time subscribes to the joint control information of the control module and sends it to the exoskeleton robot.
4. The ROS-based simulation control platform according to claim 1, characterized in that: During the human-machine coupled movement of the exoskeleton, the simulation control platform decomposes the movement state into: a state where the left foot is vacant and the right foot is supporting, a state where both feet are supporting, and a state where the right foot is vacant and the left foot is supporting.
5. The ROS-based simulation control platform according to claim 4, characterized in that: Single-leg support dynamics model: Where: is the sum of the kinetic energy of each member of the exoskeleton; is the sum of the gravitational potential energy of each member of the exoskeleton; is the elastic potential energy of the exoskeleton knee joint and; L is the Lagrangian function, which is equal to the difference between the total kinetic energy of the system and the gravitational potential energy and the total elastic potential energy; Characterize the generalized angular velocities of the hip, knee, and ankle joints, respectively; The generalized coordinates that characterize the sagittal and frontal plane systems, respectively; They respectively characterize the combined forces of the hip, knee and ankle joints, namely the joint motor driving force and the human-machine interaction force.
6. The ROS-based simulation control platform according to claim 5, characterized in that: The 6-link dynamic equation of the exoskeleton support phase is: In the formula, represents the inertia tensor matrix of the robot in the sagittal plane; is a 6x6 matrix of Coriolis and centrifugal forces; is a 6 x1 gravity matrix; is the elastic force matrix; , , They represent the 6x1-dimensional generalized position, generalized velocity, and generalized acceleration respectively.
7. The ROS-based simulation control platform according to claim 4, characterized in that: In the double-foot support state, the output torque of the left and right hip joints can be calculated based on the distance between the projection point of the exoskeleton's overall center of gravity on the line connecting the two feet's landing points and the two feet. , Proportional distribution: , is the total torque required by the left and right hip joints of the exoskeleton system; The other joint torques are derived from the 5-link dynamics: In the formula, is the kinetic energy of each link of the exoskeleton, including the left and right calves, left and right thighs, and torso; is the potential energy sum of each link of the exoskeleton, including the left and right calves, left and right thighs, and torso; Es is the elastic potential energy of the system knee joint; The generalized coordinates that characterize the sagittal and frontal plane systems, respectively; They respectively characterize the combined forces of the hip, knee and ankle joints, namely the joint motor driving force and the human-machine interaction force.
8. An exoskeleton robot simulation control system based on ROS, characterized in that: The exoskeleton robot simulation control system comprises: A lower exoskeleton end, a ROS-based simulation control platform as claimed in any one of claims 1 to 7; The lower exoskeleton end includes: a human posture sensing module, a joint drive motor module and an interactive force detection module, and the ROS-based simulation control platform can receive data fed back from the human posture sensing module and the interactive force detection module; The human posture sensing module includes: a six-axis inertial measurement sensor arranged on a back frame, angle sensors arranged at the knee joints of the left and right legs, and pressure sensors arranged at the soles of the left and right feet.
