In order to further improve efficiency and accuracy of the multi-service robot path planning in acomplex and uncertain environment, based on the mutual-coupled immune network planning algorithm, animproved measurement algorithm was proposed. According to the antigen information of obstacles and thetarget, situation-oriented and goal-oriented coupling immune network are defined respectively and theweight coefficient is used to control their roles in the overall behavior. In order to further improve theplanning performance of the robot, weight coefficient is defined dynamically according to the distance of therobot apart from the obstacles and the target. Thus we are able to optimize the robot’s behavior by choosingthe obstacle-avoidance behavior or the tend-to-target behavior in real time. The test results in the staticmulti-obstacle environment indicate that the algorithm proposed in this paper is more effective than otheralgorithms in literature, both in the aspects of length and smoothness, which verifies the ability of thealgorithm to improve the efficiency and precision of the path planning. And the multi-robot can successfullyavoid dynamic obstacles and achieve the targeting result according to the test results in the complex anduncertainty environment, which verifies its flexibility and robustness.
展开▼
