Robustness Of Gaits And Path Planning Of Rolling Locomotion For Spherical Dynamic Tensegrity

The dynamic tensegrity has the characteristics of high efficiency and adjustable structure,which enriches the research of tensegrity.The research of dynamic tensegrity in the field of vehicle system is an important direction of the research of tensegrity.Among the dynamic tensegrities applied in vehicle system,the spherical dynamic tensegrity has attracted much attentions due to its great performance in motion.The basic gait actuating the spherical dynamic tensegrity can be found by combining the actuating effects of the actuators.However,the basic gait could not actuate the dynamic tensegrity well because of the difference between numerical model and physical model.Therefore,the robustness of the gaits should be evaluated to find the gaits which could achieve the supposed movement.Meanwhile,how to find the optimal path from the start point to the target point is a vital problem to the application of spherical dynamic tensegrity in the field of vehicle system.This thesis carries out a deep investigation of the robustness of gaits and path planning for rolling locomotion of spherical dynamic tensegrity.In Chapter 2,the basic theory and method of dynamic tensegrity are introduced.The dynamic tensegriy should satisfy the self-stress equilibrium conditions and geometric stability conditions of the tensegrity structure.With the active control method,the tensegrity obtain the dynamic characteristic.The dynamic relaxation method and the genetic algorithm are used as the path tracking algorithm and the gait optimization algorithm for the dynamic tensegrity,respectively.And the design and optimazation method for gaits is a combination of these two methods.Taking the six-bar spherical dynamic tensegrity as the example,the typical basic gaits are obtained using the proposed method.In Chapter 3,the evaluation method for the robustness of the gaits is established.The definition of the robustness of the gaits is proposed and the corresponding evaluation index is given according to Monte Carlo Principle.The sensitivity parameters to the robustness of the gaits are analyzed with the turnaodo graph,and the sensitive parameters are selected to participate in the robustness evaluation.Joint random sampling of the sensitivity parameters is used to simulate the uncertainty of the system.The typical gaits given in Chapter 2 are evaluated for the robustness,and the gaits with higher robustness index are selected as the basic gaits for subsequent path planning.In Chapter 4,the path planning approach for rolling locomotion of spherical dynamic tensegrity is proposed.The possible movement path can be described with a directed graph.The path planning problem can be summarized as the problem of finding the shortest distance between specified points on the directed graph,which can be solved by Dijkstra algorithm.The distance and environmental factors are considered in this approach.Numerical examples are carried out with a six-bar spherical dynamic tensegrity showing the adaptability of the proposed approach.In Chapter 5,a physical prototype based on six-bar spherical dynamic tensegrity is developed Related physical experiments are carried out to verify the robustness evaluation method and the path planning approach.The results of experiments tally with that of the numerical examples well.It can be concluded that the method and approach proposed in this thesis are effective and reasonable.This thesis studies the dynamic movement problem of dynamic tensegrity,and proposed the evaluation method for the robustness of the gaits and the path planning approach for rolling locomotion of spherical dynamic tensegrity.The related research promotes the application of dynamic tensegrity in the field of vehicle systems.