Header

UZH-Logo

Maintenance Infos

Towards neuromorphic control: A spiking neural network based PID controller for UAV


Stagsted, Rasmus; Vitale, Antonio; Binz, Jonas; Renner, Alpha; Bonde Larsen, Leon; Sandamirskaya, Yulia (2020). Towards neuromorphic control: A spiking neural network based PID controller for UAV. In: Robotics: Science and Systems 2020, Virtual Conference, 12 July 2020 - 16 July 2020.

Abstract

In this work, we present a spiking neural network(SNN) based PID controller on a neuromorphic chip. On-chipSNNs are currently being explored in low-power AI applications.Due to potentially ultra-low power consumption, low latency,and high processing speed, on-chip SNNs are a promising toolfor control of power-constrained platforms, such as UnmannedAerial Vehicles (UAV). To obtain highly efficient and fast end-to-end neuromorphic controllers, the SNN-based AI architecturesmust be seamlessly integrated with motor control. Towards thisgoal, we present here the first implementation of a fully neu-romorphic PID controller. We interfaced Intel’s neuromorphicresearch chip Loihi to a UAV, constrained to a single degreeof freedom. We developed an SNN control architecture usingpopulations of spiking neurons, in which each spike carriesinformation about the measured, control, or error value, definedby the identity of the spiking neuron. Using this sparse code,we realize a precise PID controller. The P, I, and D gains of thecontroller are implemented as synaptic weights that can adaptaccording to an on-chip plasticity rule. In future work, theseplastic synapses can be used to tune and adapt the controllerautonomously.

Abstract

In this work, we present a spiking neural network(SNN) based PID controller on a neuromorphic chip. On-chipSNNs are currently being explored in low-power AI applications.Due to potentially ultra-low power consumption, low latency,and high processing speed, on-chip SNNs are a promising toolfor control of power-constrained platforms, such as UnmannedAerial Vehicles (UAV). To obtain highly efficient and fast end-to-end neuromorphic controllers, the SNN-based AI architecturesmust be seamlessly integrated with motor control. Towards thisgoal, we present here the first implementation of a fully neu-romorphic PID controller. We interfaced Intel’s neuromorphicresearch chip Loihi to a UAV, constrained to a single degreeof freedom. We developed an SNN control architecture usingpopulations of spiking neurons, in which each spike carriesinformation about the measured, control, or error value, definedby the identity of the spiking neuron. Using this sparse code,we realize a precise PID controller. The P, I, and D gains of thecontroller are implemented as synaptic weights that can adaptaccording to an on-chip plasticity rule. In future work, theseplastic synapses can be used to tune and adapt the controllerautonomously.

Statistics

Citations

Dimensions.ai Metrics

Altmetrics

Downloads

23 downloads since deposited on 16 Feb 2021
23 downloads since 12 months
Detailed statistics

Additional indexing

Item Type:Conference or Workshop Item (Paper), refereed, original work
Communities & Collections:07 Faculty of Science > Institute of Neuroinformatics
Dewey Decimal Classification:570 Life sciences; biology
Language:English
Event End Date:16 July 2020
Deposited On:16 Feb 2021 08:34
Last Modified:16 Feb 2021 20:30
Publisher:RSS
ISBN:9780992374761
OA Status:Hybrid
Free access at:Publisher DOI. An embargo period may apply.
Publisher DOI:https://doi.org/10.15607/rss.2020.xvi.074

Download

Hybrid Open Access

Download PDF  'Towards neuromorphic control: A spiking neural network based PID controller for UAV'.
Preview
Content: Published Version
Filetype: PDF
Size: 5MB
View at publisher