Path planning for multi-robot systems with complex tasks and time windows
Aiming at the path planning problem for multi-robot systems under the dual constraints of complex tasks and time windows,an improved ant colony algorithm is developed to find the shortest path that satisfies both the trajectory requirements represented by Boolean formulas and the time window.Firstly,based on the global map information and Boolean formulas,a preprocessing algorithm based on the Dijkstra algorithm is introduced to calculate the shortest path between task areas.Secondly,an improved ant colony algorithm based on the A∗ algorithm is proposed,incorporating the ant retreat mechanism and the heuristic function from the A∗ algorithm to enhance the solution quality and to accelerate the computation.The developed approach can find the shortest path trajectory for each robot while satisfy both the time window and logic constraints.Finally,the effectiveness of the proposed algorithm is validated through several simulation experiments.When compared to existing methods,our approach is superior in terms of solution quality and convergence properties.
multi-robot systemBoolean formulapath planningtime windows
万方数据
维普
