TY - JOUR
T1 - Neuromorphic NEF-Based Inverse Kinematics and PID Control
AU - Zaidel, Yuval
AU - Shalumov, Albert
AU - Volinski, Alex
AU - Supic, Lazar
AU - Ezra Tsur, Elishai
N1 - Publisher Copyright:
© Copyright © 2021 Zaidel, Shalumov, Volinski, Supic and Ezra Tsur.
PY - 2021/2/3
Y1 - 2021/2/3
N2 - Neuromorphic implementation of robotic control has been shown to outperform conventional control paradigms in terms of robustness to perturbations and adaptation to varying conditions. Two main ingredients of robotics are inverse kinematic and Proportional–Integral–Derivative (PID) control. Inverse kinematics is used to compute an appropriate state in a robot's configuration space, given a target position in task space. PID control applies responsive correction signals to a robot's actuators, allowing it to reach its target accurately. The Neural Engineering Framework (NEF) offers a theoretical framework for a neuromorphic encoding of mathematical constructs with spiking neurons for the implementation of functional large-scale neural networks. In this work, we developed NEF-based neuromorphic algorithms for inverse kinematics and PID control, which we used to manipulate 6 degrees of freedom robotic arm. We used online learning for inverse kinematics and signal integration and differentiation for PID, offering high performing and energy-efficient neuromorphic control. Algorithms were evaluated in simulation as well as on Intel's Loihi neuromorphic hardware.
AB - Neuromorphic implementation of robotic control has been shown to outperform conventional control paradigms in terms of robustness to perturbations and adaptation to varying conditions. Two main ingredients of robotics are inverse kinematic and Proportional–Integral–Derivative (PID) control. Inverse kinematics is used to compute an appropriate state in a robot's configuration space, given a target position in task space. PID control applies responsive correction signals to a robot's actuators, allowing it to reach its target accurately. The Neural Engineering Framework (NEF) offers a theoretical framework for a neuromorphic encoding of mathematical constructs with spiking neurons for the implementation of functional large-scale neural networks. In this work, we developed NEF-based neuromorphic algorithms for inverse kinematics and PID control, which we used to manipulate 6 degrees of freedom robotic arm. We used online learning for inverse kinematics and signal integration and differentiation for PID, offering high performing and energy-efficient neuromorphic control. Algorithms were evaluated in simulation as well as on Intel's Loihi neuromorphic hardware.
KW - Loihi
KW - neural engineering framework
KW - neuromorphic engineering
KW - robotic arm
KW - robotic control software
KW - spiking neural networks
UR - http://www.scopus.com/inward/record.url?scp=85101066337&partnerID=8YFLogxK
U2 - 10.3389/fnbot.2021.631159
DO - 10.3389/fnbot.2021.631159
M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???
C2 - 33613225
AN - SCOPUS:85101066337
SN - 1662-5218
VL - 15
JO - Frontiers in Neurorobotics
JF - Frontiers in Neurorobotics
M1 - 631159
ER -