{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T05:03:12Z","timestamp":1785301392257,"version":"3.55.0"},"reference-count":59,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T00:00:00Z","timestamp":1785283200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Robot. AI"],"abstract":"<jats:p>Reliable fault detection in multi-robot systems requires models capable of capturing complex, time-dependent fault signatures that manifest over extended temporal horizons rather than instantaneous observations alone. Existing data-driven approaches operate reactively on behavioural snapshots, failing to capture fault modes whose discriminative signature depends on temporally ordered precursors. This work formalises a theoretical impossibility result demonstrating that memoryless classifiers are provably insufficient, and presents a Long Short Term Memory (LSTM) model operating on generic, primitive behavioural features to learn these latent temporal dependencies without domain-specific priors. Two complex fault types are evaluated: progressive actuator degradation and intermittent actuator dropout, each paired with a distinct swarm movement pattern representing varying levels of temporal complexity. Experimental results demonstrate that sequential models are essential when the fault signature is defined by ordered precursors, while memoryless models partially suffice for simpler temporal structures. Compute benchmarks confirm real-time feasibility for distributed onboard deployment.<\/jats:p>","DOI":"10.3389\/frobt.2026.1827739","type":"journal-article","created":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T04:41:39Z","timestamp":1785300099000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Learning faults in time: sequential behavioural modelling for complex fault detection in multi-robot systems"],"prefix":"10.3389","volume":"13","author":[{"given":"Faisal Firas","family":"Mazloum","sequence":"first","affiliation":[{"name":"Ingeniarius, Lda.","place":["Porto, Portugal"]},{"name":"Department of Electrical and Computer Engineering, Institute of Systems and Robotics (ISR), University of Coimbra","place":["Coimbra, Portugal"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"David","family":"Portugal","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, Institute of Systems and Robotics (ISR), University of Coimbra","place":["Coimbra, Portugal"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Micael S.","family":"Couceiro","sequence":"additional","affiliation":[{"name":"Ingeniarius, Lda.","place":["Porto, Portugal"]},{"name":"Department of Electrical and Computer Engineering, Institute of Systems and Robotics (ISR), University of Coimbra","place":["Coimbra, Portugal"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,29]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","first-page":"71","DOI":"10.1016\/j.procir.2014.07.015","article-title":"Study on intermittent faults and electrical continuity","volume":"22","author":"Ahmad","year":"2014","journal-title":"Procedia Cirp"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1798","DOI":"10.1109\/TPAMI.2013.50","article-title":"Representation learning: a review and new perspectives","volume":"35","author":"Bengio","year":"2013","journal-title":"IEEE Transactions Pattern Analysis Machine Intelligence"},{"key":"B3","doi-asserted-by":"crossref","first-page":"431","DOI":"10.1007\/978-3-642-32723-0_31","article-title":"On fault tolerance and scalability of swarm robotic systems","volume-title":"Distributed Autonomous Robotic Systems: The 10th International Symposium","author":"Bjerknes","year":"2013"},{"key":"B4","doi-asserted-by":"crossref","first-page":"771","DOI":"10.1007\/978-3-031-37963-5_53","article-title":"A comparison of lstm and gru networks for learning symbolic sequences","volume-title":"Intelligent Computing","author":"Cahuantzi","year":"2023"},{"key":"B5","first-page":"253","article-title":"A distributed approach for fault detection in swarms of robots","volume-title":"Proceedings of the 23rd International Conference on Autonomous Agents and Multiagent Systems","author":"Carminati","year":"2024"},{"key":"B6","doi-asserted-by":"publisher","first-page":"72","DOI":"10.1016\/j.neucom.2018.03.067","article-title":"Time series feature extraction on basis of scalable hypothesis tests (tsfresh\u2013a python package)","volume":"307","author":"Christ","year":"2018","journal-title":"Neurocomputing"},{"key":"B7","first-page":"4006","article-title":"Failure diagnosis of discrete event systems: the case of intermittent faults","volume-title":"Proceedings of the 41st IEEE Conference on Decision and Control","author":"Contant","year":"2002"},{"key":"B8","doi-asserted-by":"crossref","DOI":"10.5772\/9507","article-title":"Diagnosis of intermittent faults and its dynamics","volume-title":"Factory Automation (Intechopen)","author":"Correcher","year":"2010"},{"key":"B9","doi-asserted-by":"publisher","first-page":"e3001","DOI":"10.7717\/peerj-cs.3001","article-title":"Deep learning in time series forecasting with transformer models and rnns","volume":"11","author":"Dos Santos","year":"2025","journal-title":"PeerJ Comput. Sci."},{"key":"B10","doi-asserted-by":"publisher","first-page":"187","DOI":"10.1016\/0167-2789(86)90240-x","article-title":"The immune system, adaptation, and machine learning","volume":"22","author":"Farmer","year":"1986","journal-title":"Phys. D. Nonlinear Phenom."},{"key":"B11","doi-asserted-by":"publisher","first-page":"349","DOI":"10.1080\/15397734.2012.670098","article-title":"Modeling and analysis of vibration-induced changes in connector resistance of high power electrical connectors for hybrid vehicles","volume":"40","author":"Fu","year":"2012","journal-title":"Mech. Based Design Structures Machines"},{"key":"B12","doi-asserted-by":"publisher","first-page":"238","DOI":"10.3390\/machines9100238","article-title":"Bearing remaining useful life prediction based on a scaled health indicator and a lstm model with attention mechanism","volume":"9","author":"Gao","year":"2021","journal-title":"Machines"},{"key":"B13","volume-title":"A Comparative Study of Detecting Anomalies in Time Series Data Using Lstm and Tcn Models","author":"Gopali","year":"2021"},{"key":"B14","doi-asserted-by":"crossref","DOI":"10.36001\/phmap.2017.v1i1.1846","article-title":"A study of the degradation of electronic speed controllers for brushless dc motors","volume-title":"Asia Pacific Conference of the Prognostics and Health Management Society 2017","author":"Gorospe","year":"2017"},{"key":"B15","doi-asserted-by":"publisher","first-page":"507","DOI":"10.48550\/arXiv.2207.08815","article-title":"Why do tree-based models still outperform deep learning on typical tabular data?","volume":"35","author":"Grinsztajn","year":"2022","journal-title":"Adv. Neural Information Processing Systems"},{"key":"B16","doi-asserted-by":"publisher","first-page":"5942","DOI":"10.3390\/s25195942","article-title":"Advancements in induction motor fault diagnosis and condition monitoring: a comprehensive review","volume":"25","author":"Hamani","year":"2025","journal-title":"Sensors Basel, Switz."},{"key":"B17","doi-asserted-by":"crossref","DOI":"10.1109\/ICMLA66185.2025.00027","article-title":"Mental multi-class classification on social media: benchmarking transformer architectures against lstm models","author":"Hasan","year":"2025"},{"key":"B18","volume-title":"Temporal cross-validation Impacts Multivariate Time Series Subsequence Anomaly Detection Evaluation","author":"Hespeler","year":"2025"},{"key":"B19","doi-asserted-by":"publisher","first-page":"1735","DOI":"10.1162\/neco.1997.9.8.1735","article-title":"Long short-term memory","volume":"9","author":"Hochreiter","year":"1997","journal-title":"Neural Comput. MIT-Press"},{"key":"B20","doi-asserted-by":"publisher","first-page":"1353879","DOI":"10.3389\/fnbot.2024.1353879","article-title":"Velocity-aware spatial-temporal attention lstm model for inverse dynamic model learning of manipulators","volume":"18","author":"Huang","year":"2024","journal-title":"Front. Neurorobotics"},{"key":"B21","doi-asserted-by":"crossref","first-page":"6875","DOI":"10.1109\/ICRA46639.2022.9812417","article-title":"Ros2swarm-a ros 2 package for swarm robot behaviors","volume-title":"2022 International Conference on Robotics and Automation (ICRA)","author":"Kaiser","year":"2022"},{"key":"B22","doi-asserted-by":"publisher","first-page":"193","DOI":"10.1016\/j.robot.2017.06.002","article-title":"Monitoring a robot swarm using a data-driven fault detection approach","volume":"97","author":"Khaldi","year":"2017","journal-title":"Robotics Aut. Syst."},{"key":"B23","article-title":"Detection of intermittent faults in aircraft electrical wire by utilizing power line communication","volume-title":"In the 9th Joint FAA\/DOD\/NASA Conference on Aging Aircraft","author":"Kim","year":"2006"},{"key":"B24","doi-asserted-by":"publisher","first-page":"1021","DOI":"10.1016\/j.robot.2011.08.008","article-title":"Adaptive data-driven error detection in swarm robotics with statistical classifiers","volume":"59","author":"Lau","year":"2011","journal-title":"Robotics Aut. Syst."},{"key":"B25","doi-asserted-by":"publisher","first-page":"011202","DOI":"10.1115\/1.4063266","article-title":"A comparative study of lstm and temporal convolutional network models for semisubmersible platform wave runup prediction","volume":"147","author":"Li","year":"2025","journal-title":"J. Offshore Mech. Arct. Eng."},{"key":"B26","doi-asserted-by":"publisher","first-page":"259","DOI":"10.1002\/nav.21947","article-title":"Health assessment and prognostics based on higher-order hidden semi-markov models","volume":"68","author":"Liao","year":"2021","journal-title":"Nav. Res. Logist. (NRL)"},{"key":"B27","doi-asserted-by":"publisher","first-page":"20200209","DOI":"10.1098\/rsta.2020.0209","article-title":"Time-series forecasting with deep learning: a survey","volume":"379","author":"Lim","year":"2021","journal-title":"Philosophical Transactions Royal Society a Mathematical, Physical Engineering Sciences"},{"key":"B28","doi-asserted-by":"publisher","first-page":"11711","DOI":"10.48550\/arXiv.2007.10319","article-title":"Mcunet: tiny deep learning on iot devices","volume":"33","author":"Lin","year":"2020","journal-title":"Adv. Neural Information Processing Systems"},{"key":"B29","volume-title":"A Critical Review of Recurrent Neural Networks for Sequence Learning","author":"Lipton","year":"2015"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.36001\/phmconf.2012.v4i1.2165","article-title":"Lithium-ion battery remaining useful life estimation based on nonlinear ar model combined with degradation feature","volume":"4","author":"Liu","year":"2012","journal-title":"Annu. Conf PHM Soc."},{"key":"B31","article-title":"Improving the performance of the lstm and hmm model via hybridization","author":"Liu","year":"2019"},{"key":"B32","volume-title":"catch22: Canonical time-series Characteristics","author":"Lubba","year":"2019"},{"key":"B33","doi-asserted-by":"publisher","first-page":"370","DOI":"10.1109\/6144.946482","article-title":"Intermittency phenomena in electrical connectors","volume":"24","author":"Maul","year":"2002","journal-title":"IEEE Trans. Components Packag. Technol."},{"key":"B34","doi-asserted-by":"publisher","first-page":"43","DOI":"10.1016\/j.jksues.2019.12.005","article-title":"A survey of modern exogenous fault detection and diagnosis methods for swarm robotics","volume":"33","author":"Miller","year":"2021","journal-title":"J. King Saud University-Engineering Sci."},{"key":"B35","doi-asserted-by":"publisher","first-page":"251252","DOI":"10.1098\/rsos.251252","article-title":"Detecting and diagnosing faults in autonomous robot swarms with an artificial antibody population model","volume":"12","author":"O\u2019Keeffe","year":"2025","journal-title":"R. Soc. Open Sci."},{"key":"B36","doi-asserted-by":"crossref","first-page":"393","DOI":"10.1007\/978-3-319-64107-2_31","article-title":"Towards fault diagnosis in robot swarms: an online behaviour characterisation approach","volume-title":"Conference Towards Autonomous Robotic Systems","author":"O\u2019Keeffe","year":"2017"},{"key":"B37","doi-asserted-by":"publisher","first-page":"131","DOI":"10.3389\/frobt.2018.00131","article-title":"Adaptive online fault diagnosis in autonomous robot swarms","volume":"5","author":"O\u2019Keeffe","year":"2018","journal-title":"Front. Robotics AI"},{"key":"B38","doi-asserted-by":"publisher","first-page":"50","DOI":"10.1016\/j.ins.2017.05.038","article-title":"A hidden markov model with dependence jumps for predictive modeling of multidimensional time-series","volume":"412","author":"Petropoulos","year":"2017","journal-title":"Inf. Sci."},{"key":"B39","doi-asserted-by":"publisher","first-page":"1066","DOI":"10.1016\/j.jsv.2005.03.007","article-title":"Wavelet filter-based weak signature detection method and its application on rolling element bearing prognostics","volume":"289","author":"Qiu","year":"2006","journal-title":"J. Sound Vibration"},{"key":"B40","first-page":"265","article-title":"A comparative study of lstm efficiency vs. transformer power for localized time series forecasting","volume-title":"2025 20th Conference on Computer Science and Intelligence Systems (Fedcsis)","author":"Sarkar","year":"2025"},{"key":"B41","doi-asserted-by":"publisher","first-page":"1101","DOI":"10.4271\/2008-01-2932","article-title":"Intermittent fault location on live electrical wiring systems","volume":"1","author":"Smith","year":"2008","journal-title":"SAE Int. J. Aerosp."},{"key":"B42","doi-asserted-by":"publisher","first-page":"1479","DOI":"10.3182\/20060829-4-cn-2909.00246","article-title":"Intermittent fault detection through message exchanges: a coherence based approach","volume":"39","author":"Soldani","year":"2006","journal-title":"IFAC Proc. Vol."},{"key":"B43","doi-asserted-by":"publisher","first-page":"061901","DOI":"10.1115\/1.4047201","article-title":"Propeller performance penalty of biofouling: computational fluid dynamics prediction","volume":"142","author":"Song","year":"2020","journal-title":"J. Offshore Mech. Arct. Eng."},{"key":"B44","doi-asserted-by":"publisher","first-page":"110455","DOI":"10.1016\/j.jss.2019.110455","article-title":"Detection of intermittent faults in software programs through identification of suspicious shared variable access patterns","volume":"159","author":"Sotiropolos","year":"2020","journal-title":"J. Syst. Softw."},{"key":"B45","doi-asserted-by":"publisher","first-page":"112228","DOI":"10.1016\/j.jss.2024.112228","article-title":"A generalized, rule-based method for the detection of intermittent faults in software programs","volume":"219","author":"Sotiropoulos","year":"2025","journal-title":"J. Syst. Softw."},{"key":"B46","volume-title":"Comparative Analysis of the Hidden Markov Model and Lstm: A Simulative Approach","author":"Tadayon","year":"2020"},{"key":"B47","volume-title":"Can Recurrent Neural Networks Warp Time?","author":"Tallec","year":"2018"},{"key":"B48","doi-asserted-by":"publisher","first-page":"016014","DOI":"10.1088\/1748-3190\/10\/1\/016014","article-title":"To err is robotic, to tolerate immunological: fault detection in multirobot systems","volume":"10","author":"Tarapore","year":"2015","journal-title":"Bioinspiration & Biomimetics"},{"key":"B49","doi-asserted-by":"publisher","first-page":"e0182058","DOI":"10.1371\/journal.pone.0182058","article-title":"Generic, scalable and decentralized fault detection for robot swarms","volume":"12","author":"Tarapore","year":"2017","journal-title":"PloS One"},{"key":"B50","doi-asserted-by":"publisher","first-page":"1516","DOI":"10.1109\/tro.2019.2929015","article-title":"Fault detection in a swarm of physical robots based on behavioral outlier detection","volume":"35","author":"Tarapore","year":"2019","journal-title":"IEEE Trans. Robotics"},{"key":"B51","doi-asserted-by":"publisher","first-page":"110883","DOI":"10.1016\/j.oceaneng.2022.110883","article-title":"Numerical methods for monitoring and evaluating the biofouling state and effects on vessels\u2019 hull and propeller performance: a review","volume":"251","author":"Valchev","year":"2022","journal-title":"Ocean. Eng."},{"key":"B52","volume-title":"LSTM Networks for Detection and Classification of Anomalies in Raw Sensor Data","author":"Verner","year":"2019"},{"key":"B53","doi-asserted-by":"publisher","first-page":"401","DOI":"10.1109\/tr.2018.2882682","article-title":"A hybrid prognostics approach for estimating remaining useful life of rolling element bearings","volume":"69","author":"Wang","year":"2018","journal-title":"IEEE Trans. Reliability"},{"key":"B54","doi-asserted-by":"publisher","first-page":"163","DOI":"10.55537\/jistr.v4i3.1298","article-title":"Comparative analysis of deep learning models for wind speed prediction using lstm, tcn and rbfnn","volume":"4","author":"Wardani","year":"2025","journal-title":"J. Inf. Syst. Technol. Res."},{"key":"B55","doi-asserted-by":"publisher","first-page":"30","DOI":"10.1504\/ijmic.2006.008645","article-title":"Safety in numbers: fault-tolerance in robot swarms","volume":"1","author":"Winfield","year":"2006","journal-title":"Int. J. Model. Identif. Control"},{"key":"B56","doi-asserted-by":"publisher","first-page":"103462","DOI":"10.1016\/j.mex.2025.103462","article-title":"Performance analysis of neural network architectures for time series forecasting: a comparative study of rnn, lstm, gru, and hybrid models","volume":"15","author":"Yunita","year":"2025","journal-title":"MethodsX"},{"key":"B57","doi-asserted-by":"publisher","first-page":"2025","DOI":"10.36227\/techrxiv.176045766.60277537\/v2","article-title":"A survey on multi-robot collaboration systems: architectures, performances, and applications","author":"Zhang","year":"","journal-title":"TechRxiv"},{"key":"B58","first-page":"193","article-title":"Revisit time series classification benchmark: the impact of temporal information for classification","volume-title":"Pacific-Asia Conference on Knowledge Discovery and Data Mining","author":"Zhang","year":""},{"key":"B59","doi-asserted-by":"crossref","first-page":"4092","DOI":"10.1109\/IROS40897.2019.8967824","article-title":"Clone swarms: learning to predict and control multi-robot systems by imitation","volume-title":"2019 IEEE\/RSJ International Conference on Intelligent Robots and Systems (IROS)","author":"Zhou","year":"2019"}],"container-title":["Frontiers in Robotics and AI"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frobt.2026.1827739\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,29]],"date-time":"2026-07-29T04:42:28Z","timestamp":1785300148000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/frobt.2026.1827739\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,29]]},"references-count":59,"alternative-id":["10.3389\/frobt.2026.1827739"],"URL":"https:\/\/doi.org\/10.3389\/frobt.2026.1827739","relation":{},"ISSN":["2296-9144"],"issn-type":[{"value":"2296-9144","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,29]]},"article-number":"1827739"}}