The Research On Obstacle Avoidance Path Planning For Multi-Agent Systems

With the continuous development of society and the rapid development of science and technology, the importance of artificial intelligence is gradually reflected in the scientific research and engineering application. Among all the highly advanced technologies, the multi-agent technology is unique and has made a systematic development on its own both in theory and in practice. At the same time, the intelligence of the agent and the complexity as well as diversity of the environment provide a dramatical challenge for the research on the obstacle avoidance path planning of the multi-agent systems, and it has become one of the most popular breakthroughs and innovations in the multi-agent technology. Nowadays, the study of obstacles will not stay in regular shape any more, but transit from static condition to dynamic motion. In order to solve the problem that how to make the general adapted rules, this paper assumed that the target can be tracked by using modern communication technology, and at the same time, from the point of view of path planning, aiming at irregular obstacles, and how to realize the obstacle avoidance path planning of multi-agents has been studied:1. Obstacle avoidance path planning for single static obstacle. As to traditional multi-agent obstacle avoidance algorithms, when considering the shape of obstacles,path redundancy and high energy consumption are always present, which results in its unpopularity. Correspondingly, we put forward a kind of irregular obstacles path planning algorithm, which defines the method of automatic identification of convexes, and integrates the sub landing point method to achieve the shortest path planning, that improved the applicability of obstacle handling and intelligent obstacle avoidance path planning.2. Multiple static irregular obstacles. Considering the diversity of obstacles in reality, centralized data processing on a number of irregular obstacles has been performed by the automatic identification of convex rules, and the concept of sub landing point has been improved while increasing the auxiliary search trace lines to make instantaneous path planning, finally, in order to avoid the phenomenon of"friction" collision, the definition of the parameters such as the safety threshold, the angular velocity and the minimum movement speed has also been developed to complete the path planning for target tracking by selecting the obstacle avoidance as a priority.3. Single irregular obstacle with uniform linear motion. In order to realize the dynamic obstacle avoidance algorithm to irregular obstacles, obstacles will be used after the automatic identification of convex rules by using a combined motion mode of uniform motion; at the same time, adding collision prediction and edge motion strategy to ensure the priority of obstacle avoidance and accordingly remaking the path planning, all of which reflect the dynamic nature of the environment and the intelligence of planning.4. Test on the spot. In order to realize the comparison between the theoretical simulation and the actual motion path, the test on the spot of the intelligent vehicle is carried out. The feasibility and effectiveness of the algorithm is verified by the comparison between the two algorithms.