Reinforcement Learning for Bio-Inspired Target Seeking

James Gillespie; Ignacio Rano; Nazmul Siddique; Jose Santos; Mehdi Khamassi

doi:10.1007/978-3-319-64107-2_52

Reinforcement Learning for Bio-Inspired Target Seeking

James Gillespie, Ignacio Rano, Nazmul Siddique, Jose Santos, Mehdi Khamassi

Research output: Chapter in Book/Report/Conference proceeding › Chapter › peer-review

2 Citations (Scopus)

Abstract

Because animals are extremely effective at moving in their natural environments they represent an excellent model to implement robust robotic movement and navigation. Braitenberg vehicles are bio- inspired models of animal navigation widely used in robotics. Tuning the parameters of these vehicles to generate appropriate behaviour can be challenging and time consuming. In this paper we present a Reinforce- ment Learning methodology to learn the sensori-motor connection of Braitenberg vehicle 3a, a biological model of source seeking. We present simulations of different stimuli and reward functions to illustrate the feasibility of this approach.

Original language	English
Title of host publication	Towards Autonomous Robotic Systems: 18th Annual Conference, TAROS 2017, Guildford, UK, July 19--21, 2017, Proceedings
Publisher	Springer
Pages	637-650
ISBN (Print)	978-3-319-64107-2
DOIs	https://doi.org/10.1007/978-3-319-64107-2_52
Publication status	E-pub ahead of print - 20 Jul 2017

Keywords

Braitenberg vehicles
Reinforcement learning
Source seeking

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1007/978-3-319-64107-2_52

Cite this

@inbook{d06539c2d7ca4203ad14dc435c0a4792,

title = "Reinforcement Learning for Bio-Inspired Target Seeking",

abstract = "Because animals are extremely effective at moving in their natural environments they represent an excellent model to implement robust robotic movement and navigation. Braitenberg vehicles are bio- inspired models of animal navigation widely used in robotics. Tuning the parameters of these vehicles to generate appropriate behaviour can be challenging and time consuming. In this paper we present a Reinforce- ment Learning methodology to learn the sensori-motor connection of Braitenberg vehicle 3a, a biological model of source seeking. We present simulations of different stimuli and reward functions to illustrate the feasibility of this approach.",

keywords = "Braitenberg vehicles, Reinforcement learning, Source seeking",

author = "James Gillespie and Ignacio Rano and Nazmul Siddique and Jose Santos and Mehdi Khamassi",

year = "2017",

month = jul,

day = "20",

doi = "10.1007/978-3-319-64107-2_52",

language = "English",

isbn = "978-3-319-64107-2",

pages = "637--650",

booktitle = "Towards Autonomous Robotic Systems: 18th Annual Conference, TAROS 2017, Guildford, UK, July 19--21, 2017, Proceedings",

publisher = "Springer",

address = "Switzerland",

}

TY - CHAP

T1 - Reinforcement Learning for Bio-Inspired Target Seeking

AU - Gillespie, James

AU - Rano, Ignacio

AU - Siddique, Nazmul

AU - Santos, Jose

AU - Khamassi, Mehdi

PY - 2017/7/20

Y1 - 2017/7/20

N2 - Because animals are extremely effective at moving in their natural environments they represent an excellent model to implement robust robotic movement and navigation. Braitenberg vehicles are bio- inspired models of animal navigation widely used in robotics. Tuning the parameters of these vehicles to generate appropriate behaviour can be challenging and time consuming. In this paper we present a Reinforce- ment Learning methodology to learn the sensori-motor connection of Braitenberg vehicle 3a, a biological model of source seeking. We present simulations of different stimuli and reward functions to illustrate the feasibility of this approach.

AB - Because animals are extremely effective at moving in their natural environments they represent an excellent model to implement robust robotic movement and navigation. Braitenberg vehicles are bio- inspired models of animal navigation widely used in robotics. Tuning the parameters of these vehicles to generate appropriate behaviour can be challenging and time consuming. In this paper we present a Reinforce- ment Learning methodology to learn the sensori-motor connection of Braitenberg vehicle 3a, a biological model of source seeking. We present simulations of different stimuli and reward functions to illustrate the feasibility of this approach.

KW - Braitenberg vehicles

KW - Reinforcement learning

KW - Source seeking

U2 - 10.1007/978-3-319-64107-2_52

DO - 10.1007/978-3-319-64107-2_52

M3 - Chapter

SN - 978-3-319-64107-2

SP - 637

EP - 650

BT - Towards Autonomous Robotic Systems: 18th Annual Conference, TAROS 2017, Guildford, UK, July 19--21, 2017, Proceedings

PB - Springer

ER -

Reinforcement Learning for Bio-Inspired Target Seeking

Abstract

Keywords

UN SDGs

Access to Document

Fingerprint

Cite this