Research On Leg Compliance Of Multi-legged Walking Robot Based On Impedance Control

In recent years, multi-legged walking robot has become a hot research topic for its good obstacle avoidance capability and strong adaptability to walk on the complex surface. In theory, good environmental adaptability and motion flexibility make sure that multi-legged walking robot will have broad application prospects in the future. Since its flexible motion capability comes from a lot of freedoms of the leg, compliance control of the leg becomes a key point for smoothly walking and tasks complete, and it is also the premise of fast walking, trotting and even running. However, the current walking robot only has poor ability of leg foot force tracking and leg compliance control, and it is difficult to achieve complex tasks in an unknown environment, which greatly affected its promotion. To solve this problem, an optimal design method of the leg based on energy consumption and target workspace is proposed and a compliance control strategy based on adaptive-fuzzy control is proposed and related experiments are carried out and analyzed on the walking robot platform.The First Chapter generally performs main applications and prospects of the multi-legged walking robot and proposes the significance of the project. On this basis, a detailed description of multi-legged walking robot development process and progress of domestic and foreign is shown. And the current status of compliance control of robot technology is carried out. In addition, compliance control of multi-legged walking robot in domestic and foreign is especially illustrated, then, research focuses of this area are analyzed. At the end, main contents of this subject are proposed.In the Second Chapter, a robot platform for the leg compliance control is built, while hardware configuration and software architecture are described in detail. Both kinematic model of the leg and equivalent model between the foot and the environment are built. By kinematics analysis, the forward/inverse kinematics of the robot foot in the joint space is solved. Additional, Least squares support vector machine (LS-SVM) based trajectory amendment algorithm is proposed according to shortcomings of currant amendment algorithms.In the Third Chapter, considering the problem of unnecessary leg dimension and energy consumption caused by the redundancy of the actual motion space of hexapod robot, an optimal design of the three joint structured leg based on the energy consumption and workspace is raised. A mathematics model of nonlinear programming with inequality constraints is established, and Augmented Lagrangian Genetic Algorithm is used to find out the globally optimal solution. Through the comparison, the structure dimension and joint rotation scope is dwindled obviously and the redundancy of the motion space is improved. The result of the optimization can satisfy the demanding of the actual tasks and with the motion requirement satisfied, the total energy consumption can be reduced.The Forth Chapter raises a control strategy based on adaptive control algorithm for compliance control of hexapod robot. The model of robot structure and impedance control is established and the indirect adaptive control algorithm is proposed. By using Lyapunov stability theorem, environmental parameters estimation is realized, and then tracking of the expected force is achieved. Through analysis of simulations and experiments, the expected force can be rapidly estimated in the practical application. It can be indicated that not only the closed-loop control of the tip force can be realized, but also smaller contact impact can be guaranteed.In Chapter5, a control strategy based on adaptive-fuzzy control algorithm for compliance control of hexapod robot under different environments is raised. According to the analysis of its parameters, the fuzzy control algorithm is used to modify the parameters of the adaptive control. Thus, the satisfied system response can be obtained through the real time adjustment according to the difference between input and output. In this chapter, the stability of impedance control and robustness of the adaptive control are also analyzed. And that environment position deviation bounds to maintain system stable is given. Comparative analysis shows that not only desired contact force can be reached when the environmental parameters is changing, but also smaller contact impact and high steady-state accuracy can be guaranteed under the height fluctuations of the body. Based on the platform, experiments of static and dynamic force control, foot force tracking under different control algorithms and different environments are implemented. Thus, the aforementioned theory and simulation are proved by the experimental results.