Roadmap Based Pursuit-Evasion and Collision Avoidance

Ship collision-avoidance and pursuit-evasion differential games with speed-loss in a turn

Computers & Mathematics with Applications ◽

10.1016/0898-1221(89)90126-0 ◽

1989 ◽

Vol 18 (1-3) ◽

pp. 77-100 ◽

Cited By ~ 17

Author(s):

T. Miloh ◽

M. Pachter

Keyword(s):

Differential Games ◽

Collision Avoidance ◽

Pursuit Evasion ◽

Ship Collision

Download Full-text

A2CM: a new multi-agent algorithm

ACTA IMEKO ◽

10.21014/acta_imeko.v10i3.1023 ◽

2021 ◽

Vol 10 (3) ◽

pp. 28

Author(s):

Gabor Paczolay ◽

Istvan Harmati

Keyword(s):

Artificial Intelligence ◽

Neural Networks ◽

Reinforcement Learning ◽

Collision Avoidance ◽

Multiple Agents ◽

Pursuit Evasion ◽

The World ◽

Multi Agent ◽

New Algorithms ◽

Modified Algorithm

<p class="Abstract">Reinforcement learning is currently one of the most researched fields of artificial intelligence. New algorithms are being developed that use neural networks to compute the selected action, especially for deep reinforcement learning. One subcategory of reinforcement learning is multi-agent reinforcement learning, in which multiple agents are present in the world. As it involves the simulation of an environment, it can be applied to robotics as well. In our paper, we use our modified version of the advantage actor–critic (A2C) algorithm, which is suitable for multi-agent scenarios. We test this modified algorithm on our testbed, a cooperative–competitive pursuit–evasion environment, and later we address the problem of collision avoidance.</p>

Download Full-text

A Simplified Pursuit-evasion Game with Reinforcement Learning

Periodica Polytechnica Electrical Engineering and Computer Science ◽

10.3311/ppee.16540 ◽

2021 ◽

Vol 65 (2) ◽

pp. 160-166

Author(s):

Gabor Paczolay ◽

Istvan Harmati

Keyword(s):

Linear Programming ◽

Reinforcement Learning ◽

Programming Problem ◽

Collision Avoidance ◽

Linear Programming Problem ◽

The State ◽

State Information ◽

Pursuit Evasion ◽

Pursuit And Evasion ◽

Evasion Game

In this paper we visit the problem of pursuit and evasion and specifically, the collision avoidance during the problem. Two distinct tasks are visited: the first is a scenario when the agents can communicate with each other online, meanwhile in the second scenario they have to only rely on the state information and the knowledge about other agents' actions. We propose a method combining the already existing Minimax-Q and Nash-Q algorithms to provide a solution that can better take the enemy as well as friendly agents' actions into consideration. This combination is a simple weighting of the two algorithms with the Minimax-Q algorithm being based on a linear programming problem.

Download Full-text