Incremental Sampling-Based Algorithms for a Class of Pursuit-Evasion Games
نویسندگان
چکیده
Pursuit-evasion games have been used for modeling various forms of conflict arising between two agents modeled as dynamical systems. Although analytical solutions of some simple pursuit-evasion games are known, most interesting instances can only be solved using numerical methods requiring significant offline computation. In this paper, a novel incremental sampling-based algorithm is presented to compute optimal open-loop solutions for the evader, assuming worst-case behavior for the pursuer. It is shown that the algorithm has probabilistic completeness and soundness guarantees. As opposed to many other numerical methods tailored to solve pursuit-evasion games, incremental sampling-based algorithms offer anytime properties, which allow their real-time implementations in online settings.
منابع مشابه
Anytime algorithms for multi-agent visibility-based pursuit-evasion games
We investigate algorithms for playing multi-agent visibilitybased pursuit-evasion games. A team of pursuers attempts to maintain visibility contact with an evader who actively avoids tracking. We aim for applicability of the algorithms in real-world scenarios; hence, we impose hard constraints on the run-time of the algorithms and we evaluate them in a simulation model based on a real-world urb...
متن کاملA Framework for Pursuit Evasion Games in R
We present a framework for solving pursuit evasion games in Rn for the case of N pursuers and a single evader. We give two algorithms that capture the evader in a number of steps linear in the original pursuer-evader distances. We also show how to generalize our results to a convex playing field with finitely many hyperplane boundaries that serve as obstacles.
متن کاملVision-based Detection of Autonomous Vehicles for Pursuit-evasion Games
We present a vision-based algorithm for the detection of multiple autonomous vehicles for a pursuit-evasion game scenario. Our algorithm computes estimates of the pose of multiple moving evaders from visual information collected by multiple moving pursuers, without previous knowledge of the segmentation of the image measurements or the number of moving evaders. We evaluate our algorithm in purs...
متن کاملMulti - Cumulant and Non - Inferior Strategies for Multi - Player
The paper presents an extension of cost-cumulant control theory over a finite horizon for a class of two-team pursuit-evasion games wherein the evolution of the states of the game in response to decision strategies selected by pursuit and evasion teams from non-inferior sets of admissible controls is described by stochastic linear differential equation and integral quadratic cost. Since the sum...
متن کاملMemory Based Learning of Pursuit Games
Combining diierent machine learning algorithms in the same system can produce beneets above and beyond what either method could achieve alone. This paper demonstrates that memory based learning can be used in conjunction with genetic algorithms to solve a diicult class of delayed reinforcement learning problems that both methods have trouble solving individually. This class, the class of diiere...
متن کامل