Intelligent agent cluster collaborative movement method and system

Abstract

The invention discloses an agent cluster cooperative motion control method and system, and the method comprises the steps: inputting the position coordinates of a target into a pre-constructed artificial potential field gravitational field model, and carrying out the calculation to obtain the target gravitational force of an agent cluster and the front centroid gravitational force of the agent cluster; inputting the centroid position coordinates of the intelligent agent cluster, the position coordinates of the target and the distance from the obstacle to the centroid of the intelligent agent cluster into a pre-constructed artificial potential field repulsive force field model, and calculating the repulsive force between the intelligent agent cluster and the obstacle; according to the agent cluster, the target point position of the agent cluster and the obstacle position, the magnitude of resultant force borne by the corresponding agent is determined; and according to the calculated resultant force applied to the agent cluster, determining an action route, and according to the action route, enabling the agent cluster to follow the virtual navigator to move towards the target. The method has the advantages that after the front centroid is introduced into the potential field, the intelligent agent cluster is effectively prevented from falling into a local minimum value, the control capability of the intelligent agent cluster is enhanced, and a target can be tracked more effectively.

Description

technical field [0001] The invention relates to a method and system for coordinated movement of intelligent body clusters, belonging to the technical field of artificial intelligence. Background technique [0002] With the rapid development of artificial intelligence and other information technologies, UAV swarm systems have also developed rapidly. As a swarm system that emerges as a carrier of swarm intelligence, such multi-robot systems based on formation control are used in many military and civilian fields. have been widely used. At the current level of robot technology, single robots are limited in terms of information acquisition, processing and execution capabilities. For complex work tasks and changing working environments, the ability of a single robot is even more insufficient. Therefore, we consider the coordination and cooperation of multiple robots to complete tasks that are impossible or difficult for a single robot. The multi-robot coordination system is bett...

AI Technical Summary

Problems solved by technology

[0003] The advantage of the artificial potential field algorithm is that the principle and implementation are relatively simple. The disadvantage is that when it is implemented, there may be a strong local minimum point in the entire potential field. If it occurs, it will cause interference and cause the entire formation task to fail.

In addition, the entire algorithm did not take other robots into consideration when designing the potential field. If the robots are prone to collisions, it will also lead to the failure of the formation task.

Images