Multi-Point Path Planning Using Bidirectional Path Search In Static And Dynamic Environments

Abstract

A multi-point robot path planning system has been presented to improve path length, safety, and smoothness. Path representation techniques that begin coding from the start point and progress one grid at a time to the destination point are presented. The purpose of multi-goal path planning is to establish a collision-free path between a set of goal poses while minimising travelled distance, maximising path safety, and maximising path smoothness. Our algorithm guarantees that planned trajectories remain safe distances from barriers, lowering the likelihood of collisions. Dijkstra's method is one of the most basic shortest-finding algorithms. A star (A*) method is a variant of Dijkstra's shortest path first algorithm that is often utilised in heuristic-based games. The idea behind A* is to assign weight to each open node and then use a heuristic to compute the total cost from beginning to end. In robotics, A* finds the most optimal path using heuristics and cost functions. Bidirectional A* reduces computation by determining the shortest path from both the source and the destination. Bidirectional search is a search approach that uses both starting and target locations to find an initial solution. Smoothing increases path robustness.