Combining Evolutionary and Adaptive Control Strategies for Quadruped Robotic Locomotion

Elisa Massi, Lorenzo Vannucci, Ugo Albanese, Marie Claire Capolei, Alexander Vandesompele, Gabriel Urbain, Angelo Maria Sabatini, Joni Dambre, Cecilia Laschi, Silvia Tolu, Egidio Falotico

Research output: Contribution to journalJournal articleResearchpeer-review

55 Downloads (Pure)

Abstract

In traditional robotics, model-based controllers are usually needed in order to bring a robotic plant to the next desired state, but they present critical issues when the dimensionality of the control problem increases and disturbances from the external environment affect the system behavior, in particular during locomotion tasks. It is generally accepted that the motion control of quadruped animals is performed by neural circuits located in the spinal cord that act as a Central Pattern Generator and can generate appropriate locomotion patterns. This is thought to be the result of evolutionary processes that have optimized this network. On top of this, fine motor control is learned during the lifetime of the animal thanks to the plastic connections of the cerebellum that provide descending corrective inputs. This research aims at understanding and identifying the possible advantages of using learning during an evolution-inspired optimization for finding the best locomotion patterns in a robotic locomotion task. Accordingly, we propose a comparative study between two bio-inspired control architectures for quadruped legged robots where learning takes place either during the evolutionary search or only after that. The evolutionary process is carried out in a simulated environment, on a quadruped legged robot. To verify the possibility of overcoming the reality gap, the performance of both systems has been analyzed by changing the robot dynamics and its interaction with the external environment. Results show better performance metrics for the robotic agent whose locomotion method has been discovered by applying the adaptive module during the evolutionary exploration for the locomotion trajectories. Even when the motion dynamics and the interaction with the environment is altered, the locomotion patterns found on the learning robotic system are more stable, both in the joint and in the task space.
Original languageEnglish
Article number71
JournalFrontiers in Neurorobotics
Volume13
Number of pages19
ISSN1662-5218
DOIs
Publication statusPublished - 2019

Keywords

  • Evolutionary algorithm
  • Bio-inspired controller
  • Cerebellum-inspired algorithm
  • Robotic locomotion
  • Neurorobotics
  • Central pattern generator

Cite this

Massi, Elisa ; Vannucci, Lorenzo ; Albanese, Ugo ; Capolei, Marie Claire ; Vandesompele, Alexander ; Urbain, Gabriel ; Maria Sabatini, Angelo ; Dambre, Joni ; Laschi, Cecilia ; Tolu, Silvia ; Falotico, Egidio. / Combining Evolutionary and Adaptive Control Strategies for Quadruped Robotic Locomotion. In: Frontiers in Neurorobotics. 2019 ; Vol. 13.
@article{2badaabb167d4e9592e1f5658eeec4f0,
title = "Combining Evolutionary and Adaptive Control Strategies for Quadruped Robotic Locomotion",
abstract = "In traditional robotics, model-based controllers are usually needed in order to bring a robotic plant to the next desired state, but they present critical issues when the dimensionality of the control problem increases and disturbances from the external environment affect the system behavior, in particular during locomotion tasks. It is generally accepted that the motion control of quadruped animals is performed by neural circuits located in the spinal cord that act as a Central Pattern Generator and can generate appropriate locomotion patterns. This is thought to be the result of evolutionary processes that have optimized this network. On top of this, fine motor control is learned during the lifetime of the animal thanks to the plastic connections of the cerebellum that provide descending corrective inputs. This research aims at understanding and identifying the possible advantages of using learning during an evolution-inspired optimization for finding the best locomotion patterns in a robotic locomotion task. Accordingly, we propose a comparative study between two bio-inspired control architectures for quadruped legged robots where learning takes place either during the evolutionary search or only after that. The evolutionary process is carried out in a simulated environment, on a quadruped legged robot. To verify the possibility of overcoming the reality gap, the performance of both systems has been analyzed by changing the robot dynamics and its interaction with the external environment. Results show better performance metrics for the robotic agent whose locomotion method has been discovered by applying the adaptive module during the evolutionary exploration for the locomotion trajectories. Even when the motion dynamics and the interaction with the environment is altered, the locomotion patterns found on the learning robotic system are more stable, both in the joint and in the task space.",
keywords = "Evolutionary algorithm, Bio-inspired controller, Cerebellum-inspired algorithm, Robotic locomotion, Neurorobotics, Central pattern generator",
author = "Elisa Massi and Lorenzo Vannucci and Ugo Albanese and Capolei, {Marie Claire} and Alexander Vandesompele and Gabriel Urbain and {Maria Sabatini}, Angelo and Joni Dambre and Cecilia Laschi and Silvia Tolu and Egidio Falotico",
year = "2019",
doi = "10.3389/fnbot.2019.00071",
language = "English",
volume = "13",
journal = "Frontiers in Neurorobotics",
issn = "1662-5218",
publisher = "Frontiers Research Foundation",

}

Massi, E, Vannucci, L, Albanese, U, Capolei, MC, Vandesompele, A, Urbain, G, Maria Sabatini, A, Dambre, J, Laschi, C, Tolu, S & Falotico, E 2019, 'Combining Evolutionary and Adaptive Control Strategies for Quadruped Robotic Locomotion', Frontiers in Neurorobotics, vol. 13, 71. https://doi.org/10.3389/fnbot.2019.00071

Combining Evolutionary and Adaptive Control Strategies for Quadruped Robotic Locomotion. / Massi, Elisa; Vannucci, Lorenzo; Albanese, Ugo; Capolei, Marie Claire; Vandesompele, Alexander; Urbain, Gabriel; Maria Sabatini, Angelo; Dambre, Joni; Laschi, Cecilia; Tolu, Silvia; Falotico, Egidio.

In: Frontiers in Neurorobotics, Vol. 13, 71, 2019.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Combining Evolutionary and Adaptive Control Strategies for Quadruped Robotic Locomotion

AU - Massi, Elisa

AU - Vannucci, Lorenzo

AU - Albanese, Ugo

AU - Capolei, Marie Claire

AU - Vandesompele, Alexander

AU - Urbain, Gabriel

AU - Maria Sabatini, Angelo

AU - Dambre, Joni

AU - Laschi, Cecilia

AU - Tolu, Silvia

AU - Falotico, Egidio

PY - 2019

Y1 - 2019

N2 - In traditional robotics, model-based controllers are usually needed in order to bring a robotic plant to the next desired state, but they present critical issues when the dimensionality of the control problem increases and disturbances from the external environment affect the system behavior, in particular during locomotion tasks. It is generally accepted that the motion control of quadruped animals is performed by neural circuits located in the spinal cord that act as a Central Pattern Generator and can generate appropriate locomotion patterns. This is thought to be the result of evolutionary processes that have optimized this network. On top of this, fine motor control is learned during the lifetime of the animal thanks to the plastic connections of the cerebellum that provide descending corrective inputs. This research aims at understanding and identifying the possible advantages of using learning during an evolution-inspired optimization for finding the best locomotion patterns in a robotic locomotion task. Accordingly, we propose a comparative study between two bio-inspired control architectures for quadruped legged robots where learning takes place either during the evolutionary search or only after that. The evolutionary process is carried out in a simulated environment, on a quadruped legged robot. To verify the possibility of overcoming the reality gap, the performance of both systems has been analyzed by changing the robot dynamics and its interaction with the external environment. Results show better performance metrics for the robotic agent whose locomotion method has been discovered by applying the adaptive module during the evolutionary exploration for the locomotion trajectories. Even when the motion dynamics and the interaction with the environment is altered, the locomotion patterns found on the learning robotic system are more stable, both in the joint and in the task space.

AB - In traditional robotics, model-based controllers are usually needed in order to bring a robotic plant to the next desired state, but they present critical issues when the dimensionality of the control problem increases and disturbances from the external environment affect the system behavior, in particular during locomotion tasks. It is generally accepted that the motion control of quadruped animals is performed by neural circuits located in the spinal cord that act as a Central Pattern Generator and can generate appropriate locomotion patterns. This is thought to be the result of evolutionary processes that have optimized this network. On top of this, fine motor control is learned during the lifetime of the animal thanks to the plastic connections of the cerebellum that provide descending corrective inputs. This research aims at understanding and identifying the possible advantages of using learning during an evolution-inspired optimization for finding the best locomotion patterns in a robotic locomotion task. Accordingly, we propose a comparative study between two bio-inspired control architectures for quadruped legged robots where learning takes place either during the evolutionary search or only after that. The evolutionary process is carried out in a simulated environment, on a quadruped legged robot. To verify the possibility of overcoming the reality gap, the performance of both systems has been analyzed by changing the robot dynamics and its interaction with the external environment. Results show better performance metrics for the robotic agent whose locomotion method has been discovered by applying the adaptive module during the evolutionary exploration for the locomotion trajectories. Even when the motion dynamics and the interaction with the environment is altered, the locomotion patterns found on the learning robotic system are more stable, both in the joint and in the task space.

KW - Evolutionary algorithm

KW - Bio-inspired controller

KW - Cerebellum-inspired algorithm

KW - Robotic locomotion

KW - Neurorobotics

KW - Central pattern generator

U2 - 10.3389/fnbot.2019.00071

DO - 10.3389/fnbot.2019.00071

M3 - Journal article

VL - 13

JO - Frontiers in Neurorobotics

JF - Frontiers in Neurorobotics

SN - 1662-5218

M1 - 71

ER -