Multi-agent cluster obstacle avoidance method based on rolling fast expansion random tree algorithm

• Main Authors: XIONG HUAJIE, YU BAOGUO, GAO DAWEI, YI QINGWU, HE CHENGLONG, WANG QING
• Format: Patent
• Language: Chinese; English
• Published: 09.04.2024
• Subjects: CONTROLLING; PHYSICS; REGULATING; SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES

The invention discloses a multi-agent cluster obstacle avoidance method based on a rolling fast expansion random tree algorithm, and belongs to the technical field of path planning of agents. The method comprises the steps that initial information is determined according to a task execution scene, a heuristic sampling method is adopted to guide an expansion tree to expand in the direction of a target point, and cluster distance constraints are added to enable a planned path to meet cluster and obstacle avoidance requirements; enabling the obtained path to meet the dynamics requirement of the intelligent agent through path smoothing processing; and through continuous forward pushing of the rolling window, the intelligent agent cluster reaches a target point area. According to the method, an RRT algorithm is improved, a rolling window, a heuristic search strategy and extension tree node distance constraints are introduced, multi-agent cluster obstacle avoidance flight in a complex unknown scene is realized, and