A collision-free trajectory planning method for multi-agents

Abstract

The invention relates to a multi-intelligent body collision-free trajectory planning method. The multi-intelligent body collision-free trajectory planning method comprises the following steps that a projection model on the motion plane of intelligent bodies is established for each intelligent body and each obstacle, wherein the model is a circular surface under the safe radius, and the projectionof the corresponding intelligent body or the corresponding obstacle on the plane is positioned in the circle through the safety radius; the collision-free constraint condition is converted into a collision penalty term based on an external penalty function method, and a continuous monitoring objective function is incorporated, wherein the collision-free constraint condition comprises the conditionthat any two circular surfaces representing the intelligent bodies do not intersect with each other and the condition that any two circular surfaces which correspondingly represent one intelligent body and one obstacle do not intersect; and the collision-free trajectory of each intelligent body is obtained by taking the minimization objective function as a target, and by adopting an optimal control theory and a gradient descent method. According to the multi-intelligent body collision-free trajectory planning method, based on the modeling mode of the circular surface and the objective function formulation based on the external penalty function method, the dependence of the model on the intelligent bodies and obstacles of various shapes is reduced, moreover, the final optimized trajectoryis driven to be away from the occurrence of the collision phenomenon, and finally the collision-free optimal monitoring trajectory is obtained.

Description

technical field [0001] The invention belongs to the field of multi-agent continuous monitoring, and more specifically relates to a multi-agent non-collision trajectory planning method. Background technique [0002] In recent years, due to the rapid development of robot technology and sensor network technology, multi-agent collaborative network has received extensive attention. Due to its excellent characteristics such as high autonomy and strong collaboration ability, researchers have applied it to many complex and difficult comparison networks. In large tasks, such as environment sampling, object tracking, resource collection, and continuous monitoring. Continuous monitoring is to continuously monitor the state of the target of interest in the task area by controlling the movement of multiple agents with sensors with perception and communication functions to complete a variety of tasks. Different from the traditional overlay monitoring where a large number of static sensor...

AI Technical Summary

Problems solved by technology

[0005] The present invention provides a multi-agent non-collision trajectory planning method, which is used to solve the agent that is neglected due to the agent modeling method in the existing agent trajectory planning method for multi-agent continuous monitoring but may still exist in reality. Technical issues with agents and agents colliding with obstacles

Images