A multi-agent based evolutionary artificial neural network for general navigation in unknown environments

Abstract

This paper presents a multi-agent based evolutionary artificial neural network (ANN) for general navigation. While vision is a single input channel to the ANN, only the information about the availability of places in the current visual field is considered so that navigation is executed without restriction to any specific environment or object. Through constant interaction with the environment, multiple agents co-decide and compete with each other for the move decisions. These agents are subject to evolution via evolutionary strategies which are believed to be capable of solving the epistasis problem. Evolutionary learning continues during the lifetime of the ANN and is triggered whenever a correct move decision is absent. These basic ideas are illustrated by a virtual creature exploring unknown environments as far as possible with obstacle avoidance. Experimental results have shown that the virtual creature has incrementally improved its navigation ability and explored the environments efficiently. Meanwhile, some natural and robust behaviors have emerged, which are not programmed in advance.

Multi-agent navigation based on deep reinforcement learning and traditional pathfinding algorithm

“…The main inspiration of researchers is usually to make the decision policy highly adaptive to unknown or stochastic environments. Some latest progresses have been made via evolutionary artificial neural networks (ANN) [19] or reinforcement learning implemented by evolutionary neural networks [11]. Other works usually aim at coping with very large scale problems in which agents can only gather partial and local information of environment.…”

Fuzzy Counter Ant Algorithm for Maze Problem

“…For example, knowledge of an environment to be navigated is often required to be known in advance. The designer is required to revise the previous knowledge of features or map so as to correspond to the new changes in the environment [6].…”

“…6 below demonstrates the integration of various components to form a system. As shown in figure, the robot carries an ID ring over its head and has a WIFI card enabled through which it can send the sensor readings back to the Integration of Different Components in lab remote workstation, which is also connected to a WIFI router.…”

A Fuzzy Action Selection Method for Virtual Agent Navigation in Unknown Virtual Environments

“…The visual sensor in [27] has been modified by adding a method proposed in [28] which used the DempsterShafer theory of evidence [29]. Its field of vision range is 180 o and five cells in distance.…”