Collision Avoidance System Synthesis for a Group of Robots in Unsupervised Learning Paradigm
Anton Dotsenko · Mekhatronika Avtomatizatsiya Upravlenie · 2020
Collisionavoidanceis very important problem in the domain of multi-robot interaction. In this paper we propose a newapproachofcollisionavoidanceinthecontextoftheoptimalcontrolsystemsynthesisproblemdefinitionwithminimalinformationavailable.Itis assumed that robots have a certain scope within which they can interact with static and dynamic phaseconstraints.A group of robots is considered to be homogeneous, and control system unit for reaching terminal states alreadyavailabletorobots.The control system which is responsible for collision avoidance is only activated when the nearest neighboris located in the scope of the considered robot. The first important feature of this work is the fact that the collision avoidancebetweentworobotsis reciprocal with joint control system, without assigning priorities. Another key feature of this work is the complete absence of information about the environment and the current state of other robots at given time. Robots only shareinformationwithnearestneighborsiftheylocateinthescopeofeachother.Wealsopresenta computational experiment withmobilerobotsas control objects. A multilayer perceptron was used to approximate the control function. Weights of the perceptron were optimized in unsupervised paradigm by an algorithm belonging to the evolutionary strategies class. At the beginningofeachepochwegeneratea sample of collision scenarios for optimization, while the quality criterion of the achieved weights attheendof epochis evaluated on a fixed test sample. Experimental results demonstrate strong ability of the optimized multilayerperceptrontomaptherelativestateoftwomobilerobotstocontrolsinordertoavoidcollisions.