Optimizing Pathfinding in Dynamic Environments: A Comparative Study of A* and D-Lite Algorithms

Abstract

This is one of the most important challenges in domains like robotics, AI, and video games: given some start point, there generally needs to be a minimum cost way found to reach a specified goal. Algorithms like A would not change the static conditions since they compute extremely efficiently when the environments are static. But for dynamic environments, with all the new challenges of moving obstacles and changing terrains besides others, do increase the complexity. We compare to a quite heuristically well-known A* against an incremental search algorithm which we call D-Lite. Our ability is to deal with dynamic environments ubiquitously. In simulation, we demonstrate how though A* is highly suited for static environments, D-Lite best competes in dynamic scenarios owing to its ability to incrementally adapt to changing conditions without calculating paths anew from scratch. This explains why real-time applications are perfect for optimum pathfinding. It also throws much light on how D-Lite in dynamic algorithms adapts to environments for optimum operations.