Hybrid Driven Formation And Motion Control For Multi-Robot

In recent years,with the vigorous development of information technology,cybernetics,artificial intelligence theory and other multidisciplinary theories and advanced technologies,robotics technology has made great progress.Compared with the single robot system,multirobot systems exhibit better robustness and environmental adaptability,and can cooperate to complete more complex tasks.In view of these unique advantages,multi-robot systems have shown broad application prospects in many industries,such as national defense,industrial manufacturing,warehousing logistics and service industries,Thus,they have gained increasing attention from both the industry and academia.As the core issue of multi-robot coordinated control,multi-robot formation control has long been the focus for researchers.Corresponding research remains at the rudimentary stage of exploration and development.Plenty of theoretical and technical issues deserve efficient resolutions urgently.In particular,one direction is the multi-robot systems formation control under complex obstacle environment.This thesis will study the formation of multi-robot system and formation control in complex environment and propose the following solutions:(1)The single-robot motion model is established by using the improved artificial potential field method to make the single-robot move from the initial position to the target position during the process.It not only can realize real-time obstacle avoidance,but also can avoid the target unreachable problem of traditional artificial potential field.(2)Based on the mathematical programming model,an optimization algorithm for determining the target position of multi-robot is proposed.And a programming function that meets the requirements of the shortest path and minimum time consumption is constructed.We use the characteristics of the single-robot motion model and the multi-robot distributed control structure to establish the mathematical model of the repulsive force field between robots,so that the multi-robot can form a regular polygon formation according to the random initial position separately.(3)On the basis of the single-robot motion model and the formation of multi-robot formations,by combining linear temporal logic(LTL)with artificial potential field method,a discrete online flexible formation control strategy for multi-robot systems is proposed,so that multi-robots choose the appropriate formation in real time according to the actual environment during the movement,pass through complex obstacles,and restore the original formation gradually under the condition of satisfying the movement constraints of multiple robots.Finally,the algorithm proposed in this thesis is simulated and verified by MATLAB software.The simulation results show that:(1)Using the improved artificial potential field method,a single robot can effectively avoid static and dynamic obstacles and plan according to the shortest path principle.(2)On the basis of single-robot motion planning,according to the optimization algorithm for determining the target position and the inter-robot force,the multi-robots realize the formation of a regular polygon formation with any number of robots under the premise of the least energy consumption or the shortest time consumption,and realize the formation of multi-robot formations that fixes a certain robot.(3)Based on the single-robot motion planning and formation of multi-robot formations,according to the LTL method and artificial potential field method,the multi-robot systems compress formation or change formation through the communication among robots in the process of multi-robot movement in the complex obstacle environment,realizing collision avoidance,obstacle avoidance,formation and formation recovery,before reaching the corresponding target positions,which reflects the ability of multi-robot to make decision according to the actual environment independently.The above research results show that the algorithm proposed in this thesis is of the real-time and distributed characteristics.Each robot can plan motion according to the environment information obtained by itself,which improves the flexibility and applicability of the system,and solves the problem of multi-robot system being unable to pass through complex obstacles as a whole and leading to the defect of stagnation.