Conflict-Based Search
Overview
Conflict-Based Search (CBS) coordinates agents by detecting a collision, branching on constraints that prevent it, and replanning affected agents individually; under standard assumptions, this can find optimal paths. Variants such as continuous-time CBS, SIPP-based planners, and anytime or bounded-gap search extend the method to asynchronous actions, non-unit costs, and limited computation. CBS is used on standard multi-agent pathfinding benchmark maps and in settings such as warehouse fleets, elevators, and multi-robot manipulation, while conflict-directed reasoning also appears in verified search. A central challenge is retaining strong guarantees as richer dynamics and online disturbances increase the cost of conflict detection and replanning.
Latest papers
17 papers1 paper in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.