Real-Time Path Planning for Tethered Robots in Extreme Environments Based on Constraint Consistency and Optimality

The tether enables the robot to operate in extreme environments while imposing constraints on length and the maximum deployment angle in path planning. Conventional path-planning algorithms for tethered robots suffer from tight coupling between constraint evaluation and the search process, resulting in low planning efficiency. To address this issue, we propose a method based on our constraint consistency and optimality (CCO) theorem. Instead of computing tether constraints during the search for robot paths, we first search for terminal tether configurations and evaluate the constraints only afterward, effectively decoupling constraint evaluation from the search process. In addition, we introduce the Receptive Field Enhanced Dijkstra (RFE-Dijkstra) algorithm, which significantly reduces the search space and further accelerates computation. Experiments show that the proposed CCO-based planner runs in milliseconds on grid maps with dozens of obstacles, using only 0.17% of the computation time of direct shortest-path search and just 0.014% of that of the method that precomputes the reachable space before searching.