{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T22:43:34Z","timestamp":1781909014575,"version":"3.54.5"},"reference-count":62,"publisher":"IOP Publishing","issue":"1","license":[{"start":{"date-parts":[[2022,1,24]],"date-time":"2022-01-24T00:00:00Z","timestamp":1642982400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2022,1,24]],"date-time":"2022-01-24T00:00:00Z","timestamp":1642982400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/iopscience.iop.org\/info\/page\/text-and-data-mining"}],"funder":[{"DOI":"10.13039\/100000015","name":"U.S. Department of Energy","doi-asserted-by":"crossref","award":["DE-AC05-00OR22725"],"award-info":[{"award-number":["DE-AC05-00OR22725"]}],"id":[{"id":"10.13039\/100000015","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["iopscience.iop.org"],"crossmark-restriction":false},"short-container-title":["Neuromorph. Comput. Eng."],"published-print":{"date-parts":[[2022,3,1]]},"abstract":"<jats:title>Abstract<\/jats:title>\n               <jats:p>Neuromorphic computing offers the opportunity to implement extremely low power artificial intelligence at the edge. Control applications, such as autonomous vehicles and robotics, are also of great interest for neuromorphic systems at the edge. It is not clear, however, what the best neuromorphic training approaches are for control applications at the edge. In this work, we implement and compare the performance of evolutionary optimization and imitation learning approaches on an autonomous race car control task using an edge neuromorphic implementation. We show that the evolutionary approaches tend to achieve better performing smaller network sizes that are well-suited to edge deployment, but they also take significantly longer to train. We also describe a workflow to allow for future algorithmic comparisons for neuromorphic hardware on control applications at the edge.<\/jats:p>","DOI":"10.1088\/2634-4386\/ac45e7","type":"journal-article","created":{"date-parts":[[2021,12,22]],"date-time":"2021-12-22T22:41:45Z","timestamp":1640212905000},"page":"014002","update-policy":"https:\/\/doi.org\/10.1088\/crossmark-policy","source":"Crossref","is-referenced-by-count":22,"title":["Evolutionary vs imitation learning for neuromorphic control at the edge*"],"prefix":"10.1088","volume":"2","author":[{"given":"Catherine","family":"Schuman","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Robert","family":"Patton","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shruti","family":"Kulkarni","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Maryam","family":"Parsa","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Christopher","family":"Stahl","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"N Quentin","family":"Haas","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"J Parker","family":"Mitchell","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shay","family":"Snyder","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Amelie","family":"Nagle","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Alexandra","family":"Shanafield","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Thomas","family":"Potok","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"266","published-online":{"date-parts":[[2022,1,24]]},"reference":[{"key":"nceac45e7bib1","first-page":"552","article-title":"Neuromorphic self-tuning PID controller","author":"Akhyar","year":"1993"},{"key":"nceac45e7bib2","first-page":"1","article-title":"Grant: ground-roaming autonomous neuromorphic targeter","author":"Ambrose","year":"2020"},{"key":"nceac45e7bib3","first-page":"124","article-title":"A 55\u00a0nm time-domain mixed-signal neuromorphic accelerator with stochastic synapses and embedded reinforcement learning for autonomous micro-robots","author":"Amravati","year":"2018"},{"key":"nceac45e7bib4","first-page":"1614","article-title":"f1tenth.dev-an open-source ROS based f1\/10 autonomous racing simulator","author":"Babu","year":"2020"},{"key":"nceac45e7bib5","doi-asserted-by":"publisher","first-page":"1575","DOI":"10.1109\/tbcas.2019.2953001","article-title":"Real-time ultra-low power ECG anomaly detection using an event-driven neuromorphic processor","volume":"13","author":"Bauer","year":"2019","journal-title":"IEEE Trans. Biomed. Circuits Syst."},{"key":"nceac45e7bib6","first-page":"944","article-title":"Review of electrical architectures and power requirements for automated vehicles","author":"Baxter","year":"2018"},{"key":"nceac45e7bib7","first-page":"1","article-title":"Benchmarking keyword spotting efficiency on neuromorphic hardware","author":"Peter","year":"2019"},{"key":"nceac45e7bib8","doi-asserted-by":"publisher","first-page":"14","DOI":"10.1109\/mspec.2020.9126102","article-title":"Nvidia makes it easy to embed AI: the Jetson Nano packs a lot of machine-learning power into DIY projects\u2014(hands on)","volume":"57","author":"Cass","year":"2020","journal-title":"IEEE Spectr."},{"key":"nceac45e7bib9","first-page":"928","article-title":"From insight to impact: building a sustainable edge computing platform for smart homes","author":"Chang","year":"2018"},{"key":"nceac45e7bib10","doi-asserted-by":"publisher","DOI":"10.1126\/scirobotics.abb6938","article-title":"A memristor-based hybrid analog-digital computing platform for mobile robotics","volume":"5","author":"Chen","year":"2020","journal-title":"Sci. Robot."},{"key":"nceac45e7bib11","doi-asserted-by":"publisher","first-page":"34","DOI":"10.1109\/msp.2020.2985815","article-title":"Event-based neuromorphic vision for autonomous driving: a paradigm shift for bio-inspired visual sensing and perception","volume":"37","author":"Chen","year":"2020","journal-title":"IEEE Signal Process. Mag."},{"key":"nceac45e7bib12","first-page":"2884","article-title":"Deep imitation learning for autonomous driving in generic urban scenarios with enhanced safety","author":"Chen","year":"2019"},{"key":"nceac45e7bib13","first-page":"1571","article-title":"Library for evolutionary algorithms in python (LEAP)","author":"Coletti","year":"2020"},{"key":"nceac45e7bib14","doi-asserted-by":"publisher","first-page":"911","DOI":"10.1109\/jproc.2021.3067593","article-title":"Advancing neuromorphic computing with Loihi: a survey of results and outlook","volume":"109","author":"Davies","year":"2021","journal-title":"Proc. IEEE"},{"key":"nceac45e7bib15","doi-asserted-by":"publisher","first-page":"356","DOI":"10.1007\/s13177-019-00204-2","article-title":"Complex trajectory tracking using PID control for autonomous driving","volume":"18","author":"Farag","year":"2020","journal-title":"Int. J. Intell. Transp. Syst. Res."},{"key":"nceac45e7bib16","doi-asserted-by":"publisher","DOI":"10.1016\/j.adapen.2020.100006","article-title":"Smart grid encounters edge computing: opportunities and applications","volume":"1","author":"Feng","year":"2021","journal-title":"Adv. Appl. Energy"},{"key":"nceac45e7bib17","first-page":"1","article-title":"Neuromorphic self-driving robot with retinomorphic vision and spike-based processing\/closed-loop control","author":"Fischl","year":"2017"},{"key":"nceac45e7bib18","doi-asserted-by":"publisher","first-page":"3249","DOI":"10.1021\/acs.est.7b04576","article-title":"Life cycle assessment of connected and automated vehicles: sensing and computing subsystem and vehicle level effects","volume":"52","author":"Gawron","year":"2018","journal-title":"Environ. Sci. Technol."},{"key":"nceac45e7bib19","first-page":"9631","article-title":"Adaptive motor control and learning in a spiking neural network realised on a mixed-signal neuromorphic processor","author":"Glatz","year":"2019"},{"key":"nceac45e7bib20","doi-asserted-by":"publisher","first-page":"1729881419885288","DOI":"10.1177\/1729881419885288","article-title":"Three-dimensional walking of a simulated muscle-driven quadruped robot with neuromorphic two-level central pattern generators","volume":"16","author":"Habu","year":"2019","journal-title":"Int. J. Adv. Robot. Syst."},{"key":"nceac45e7bib21","doi-asserted-by":"publisher","first-page":"6239","DOI":"10.1109\/lra.2020.3012129","article-title":"Evolved neuromorphic control for high speed divergence-based landings of MAVs","volume":"5","author":"Hagenaars","year":"2020","journal-title":"IEEE Robot. Autom. Lett."},{"key":"nceac45e7bib22","first-page":"635","article-title":"A self-driving robot using deep convolutional neural networks on neuromorphic hardware","author":"Hwu","year":"2017"},{"key":"nceac45e7bib23","article-title":"BayesRace: learning to race autonomously using prior experience","author":"Jain","year":"2020"},{"key":"nceac45e7bib24","doi-asserted-by":"publisher","first-page":"49","DOI":"10.1016\/j.bica.2016.11.002","article-title":"A historical survey of algorithms and hardware architectures for neural-inspired and neuromorphic computing applications","volume":"19","author":"James","year":"2017","journal-title":"Biol. Insp. Cogn. Archit."},{"key":"nceac45e7bib25","doi-asserted-by":"publisher","first-page":"51","DOI":"10.3389\/fnbot.2020.00051","article-title":"Event-based robotic grasping detection with neuromorphic vision sensor and event-grasping dataset","volume":"14","author":"Li","year":"2020","journal-title":"Front. Neurorobot."},{"key":"nceac45e7bib26","article-title":"Federated transfer reinforcement learning for autonomous driving","author":"Liang","year":"2019"},{"key":"nceac45e7bib27","article-title":"Short-term trajectory planning using reinforcement learning within a neuromorphic control architecture","author":"Mirus","year":"2019"},{"key":"nceac45e7bib28","first-page":"136","article-title":"Neon: neuromorphic control for autonomous robotic navigation","author":"Mitchell","year":"2017"},{"key":"nceac45e7bib29","first-page":"1","article-title":"Caspian: a neuromorphic development platform","author":"Mitchell","year":"2020"},{"key":"nceac45e7bib30","first-page":"1","article-title":"A small, low cost event-driven architecture for spiking neural networks on FPGAs","author":"Mitchell","year":"2020"},{"key":"nceac45e7bib31","first-page":"77","article-title":"F1TENTH: an open-source evaluation environment for continuous control and reinforcement learning","author":"O\u2019Kelly","year":"2020"},{"key":"nceac45e7bib32","doi-asserted-by":"publisher","first-page":"286","DOI":"10.1177\/0278364919880273","article-title":"Imitation learning for agile autonomous driving","volume":"39","author":"Pan","year":"2020","journal-title":"Int. J. Robot. Res."},{"key":"nceac45e7bib33","doi-asserted-by":"crossref","DOI":"10.1145\/3477145.3477170","article-title":"Neuromorphic computing for autonomous racing","author":"Patton","year":"2021"},{"key":"nceac45e7bib34","first-page":"2825","article-title":"Scikit-learn: machine learning in python","volume":"12","author":"Pedregosa","year":"2011","journal-title":"J. Mach. Learn. Res."},{"key":"nceac45e7bib35","first-page":"57","article-title":"Autonomous driving of a rover-like robot using neuromorphic computing","author":"Pi\u00f1ero-Fuentes","year":"2021"},{"key":"nceac45e7bib36","article-title":"The TENNLab suite of LIDAR-based control applications for recurrent, spiking, neuromorphic systems","author":"Plank","year":"2019"},{"key":"nceac45e7bib37","doi-asserted-by":"publisher","first-page":"17","DOI":"10.1109\/locs.2018.2885976","article-title":"The TENNLab exploratory neuromorphic computing framework","volume":"1","author":"Plank","year":"2018","journal-title":"IEEE Lett. Comput. Soc."},{"key":"nceac45e7bib38","first-page":"1","article-title":"An astrocyte-modulated neuromorphic central pattern generator for hexapod robot locomotion on Intel\u2019s Loihi","author":"Polykretis","year":"2020"},{"key":"nceac45e7bib39","first-page":"1","article-title":"Learning first-to-spike policies for neuromorphic control using policy gradients","author":"Rosenfeld","year":"2019"},{"key":"nceac45e7bib40","first-page":"1","article-title":"Evolutionary optimization for neuromorphic systems","author":"Schuman","year":"2020"},{"key":"nceac45e7bib41","first-page":"1","article-title":"Non-traditional input encoding schemes for spiking neuromorphic systems","author":"Schuman","year":"2019"},{"key":"nceac45e7bib42","first-page":"145","article-title":"An evolutionary optimization framework for neural networks and neuromorphic architectures","author":"Schuman","year":"2016"},{"key":"nceac45e7bib43","first-page":"1","article-title":"A software framework for comparing training approaches for spiking neuromorphic systems","author":"Schuman","year":"2021"},{"key":"nceac45e7bib44","article-title":"A survey of neuromorphic computing and neural networks in hardware","author":"Schuman","year":"2017"},{"key":"nceac45e7bib45","first-page":"1","article-title":"Low size, weight, and power neuromorphic computing to improve combustion engine efficiency","author":"Schuman","year":"2020"},{"key":"nceac45e7bib46","article-title":"Avoiding excess computation in asynchronous evolutionary algorithms","author":"Scott","year":"2021"},{"key":"nceac45e7bib47","doi-asserted-by":"publisher","first-page":"86","DOI":"10.1038\/s42256-018-0015-y","article-title":"Training deep neural networks for binary communication with the Whetstone method","volume":"1","author":"Severa","year":"2019","journal-title":"Nat. Mach. Intell."},{"key":"nceac45e7bib48","doi-asserted-by":"publisher","DOI":"10.1088\/1748-3190\/ac290c","article-title":"LIDAR-driven spiking neural network for collision avoidance in autonomous driving","volume":"16","author":"Shalumov","year":"2021","journal-title":"Bioinsp. Biomim."},{"key":"nceac45e7bib49","first-page":"p 1905","article-title":"Live demonstration: spiking neural circuit based navigation inspired by C. elegans thermotaxis","author":"Shetty","year":"2015"},{"key":"nceac45e7bib50","first-page":"8992","article-title":"Formulazero: distributionally robust online adaptation via offline population synthesis","author":"Sinha","year":"2020"},{"key":"nceac45e7bib51","first-page":"46","article-title":"Neuromorphic closed-loop control of a flexible modular robot by a simulated spiking central pattern generator","author":"Spaeth","year":"2020"},{"key":"nceac45e7bib52","doi-asserted-by":"publisher","DOI":"10.15607\/rss.2020.xvi.074","article-title":"Towards neuromorphic control: a spiking neural network based PID controller for UAV","author":"Stagsted","year":"2020","journal-title":"Robotics: Sci. Syst."},{"key":"nceac45e7bib53","author":"Stewart","year":"2012"},{"key":"nceac45e7bib54","first-page":"p V003T37A012","article-title":"A fast integrated planning and control framework for autonomous driving via imitation learning","volume":"vol 51913","author":"Sun","year":"2018"},{"key":"nceac45e7bib55","doi-asserted-by":"publisher","first-page":"100","DOI":"10.1016\/j.neunet.2019.03.005","article-title":"Recent advances in physical reservoir computing: a review","volume":"115","author":"Tanaka","year":"2019","journal-title":"Neural Netw."},{"key":"nceac45e7bib56","first-page":"1","article-title":"CarSNN: an efficient spiking neural network for event-based autonomous cars on the Loihi neuromorphic research processor","author":"Viale","year":"2021"},{"key":"nceac45e7bib57","doi-asserted-by":"publisher","first-page":"1900103","DOI":"10.1002\/aisy.201900103","article-title":"A Braitenberg vehicle based on memristive neuromorphic circuits","volume":"2","author":"Wang","year":"2020","journal-title":"Adv. Intell. Syst."},{"key":"nceac45e7bib58","doi-asserted-by":"publisher","first-page":"260","DOI":"10.3389\/fnins.2019.00260","article-title":"Demonstrating advantages of neuromorphic computation: a pilot study","volume":"13","author":"Wunderlich","year":"2019","journal-title":"Front. Neurosci."},{"key":"nceac45e7bib59","first-page":"815","article-title":"Advanced motion control: from classical PID to nonlinear adaptive robust control","author":"Yao","year":"2010"},{"key":"nceac45e7bib60","doi-asserted-by":"publisher","first-page":"631159","DOI":"10.3389\/fnbot.2021.631159","article-title":"Neuromorphic NEF-based inverse kinematics and PID control","volume":"15","author":"Zaidel","year":"2021","journal-title":"Front. Neurorobot."},{"key":"nceac45e7bib61","first-page":"89","article-title":"Neuromorphic implementation of spiking relational neural network for motor control","author":"Zhao","year":"2020"},{"key":"nceac45e7bib62","doi-asserted-by":"publisher","first-page":"546","DOI":"10.1109\/jetcas.2020.3040390","article-title":"Closed-loop spiking control on a neuromorphic processor implemented on the iCub","volume":"10","author":"Zhao","year":"2020","journal-title":"IEEE J. Emerg. Sel. Top. Circuits Syst."}],"container-title":["Neuromorphic Computing and Engineering"],"original-title":[],"link":[{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ac45e7","content-type":"text\/html","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ac45e7\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ac45e7","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ac45e7\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ac45e7\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ac45e7\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ac45e7\/pdf","content-type":"application\/pdf","content-version":"am","intended-application":"similarity-checking"},{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ac45e7\/pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,1,21]],"date-time":"2022-01-21T10:18:05Z","timestamp":1642760285000},"score":1,"resource":{"primary":{"URL":"https:\/\/iopscience.iop.org\/article\/10.1088\/2634-4386\/ac45e7"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,1,24]]},"references-count":62,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2022,1,24]]},"published-print":{"date-parts":[[2022,3,1]]}},"URL":"https:\/\/doi.org\/10.1088\/2634-4386\/ac45e7","relation":{},"ISSN":["2634-4386"],"issn-type":[{"value":"2634-4386","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,1,24]]},"assertion":[{"value":"Evolutionary vs imitation learning for neuromorphic control at the edge*","name":"article_title","label":"Article Title"},{"value":"Neuromorphic Computing and Engineering","name":"journal_title","label":"Journal Title"},{"value":"paper","name":"article_type","label":"Article Type"},{"value":"\u00a9 2022 The Author(s). Published by IOP Publishing Ltd","name":"copyright_information","label":"Copyright Information"},{"value":"2021-10-06","name":"date_received","label":"Date Received","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2021-12-22","name":"date_accepted","label":"Date Accepted","group":{"name":"publication_dates","label":"Publication dates"}},{"value":"2022-01-24","name":"date_epub","label":"Online publication date","group":{"name":"publication_dates","label":"Publication dates"}}]}}