Maze Search Using Reinforcement Learning by a Mobile Robot

Journal Title: Advances in Robotics & Mechanical Engineering - Year 2018, Vol 1, Issue 2

Abstract

This review presents on research of application of reinforcement learning and new approaches on a course search in mazes with some kinds of multi-point passing as machines. It is based on a selective learning from multi-directive behavior patterns using PS (Profit Sharing) by an agent. The behavior is selected stochastically from 4 kinds of ones using PS with Boltzmann Distribution with a plan to inhibit invalid rules by a reinforcement function of a geometric sequence. Moreover, a variable temperature scheme is adopted in this distribution, where the environmental identification is valued in the first stage of the search and the convergence of learning is shifted to be valuing as time passing. A SUB learning system and a multistage layer system were proposed in this review, and these functions were inspected by some simulations and experiments using a mobile robot.In robots which has begun to spread to not only industrial world but also general home, e.g. cleaning robots etc., recently achievement of complex tasks and adaptation of complex environment has been required and can be done by agents which were concept of distributed artificial intelligent and caught abstractly various robots. Conventionally, as behavior of agents has been controlled by rules designed as if then rules, a lot of rules were required for adaptation to complex environment and achievement of complex tasks.

Authors and Affiliations

Makoto Katoh, Keiichi Tanaka, Shunsuke Shikichi

Keywords

Related Articles

Robotic Farmers in Agriculture

Revolution of Robotic farmer is on the way, fruit picking machines are ready to roll into the fields and will replace human workers at one point of time. A Robot Farmer is just a one of the new technologies that will com...

Use of Latest Robots for Autism Spectrum Disorder

One of the prominent applications of robots is in assistive therapy using humanoids. Robots are now playing a vital role in our lives as assistants, therapist, companions and much more. Autism Spectrum Disorder (ASD) aff...

Editorial on Self-Healing Robots

Scientists at University of Colorado, Boulder had developed a first and fully re-healable and recyclable electronic material. This novel technology functions like properties of human skin, capable of measuring pressure,...

Simulation of the Instability of Gas Layer Flow Inside Fluid

The stability of a thin gas layer flow between two fluid layers moving in the same or opposite direction is modeled and simulated numerically. A linear stability is considered for the system using the non-stationary equa...

Research on Fishery Water Quality Detection System of Unmanned Aerial Vehicle

Many lands in China are close to rivers and oceans and they are very suitable for developing fisheries. Water quality monitoring is indispensable to the development of fisheries. Therefore, a method of detecting, evaluat...

Download PDF file
  • EP ID EP580565
  • DOI 10.32474/ARME.2018.01.000110
  • Views 106
  • Downloads 0

How To Cite

Makoto Katoh, Keiichi Tanaka, Shunsuke Shikichi (2018). Maze Search Using Reinforcement Learning by a Mobile Robot. Advances in Robotics & Mechanical Engineering, 1(2), 34-34. https://europub.co.uk/articles/-A-580565