{"id":{"repo_id":"cork","oai_identifier":"oai:cora.ucc.ie:10468/18509"},"canonical_url":"https://search.dev.ndltd.org/etd/cork/oai:cora.ucc.ie:10468/18509","repository":{"repo_id":"cork","name":"University College Cork","base_url":"https://cora.ucc.ie/server/oai/request"},"display":{"title":"Formation control of multiple robots under packet loss","abstract":"This thesis addresses the challenges posed by communication packet loss in leader-follower formation control of multi-robot systems, with a focus on Automated Guided Vehicles (AGVs) and drones. It presents a comprehensive analysis of how packet loss affects linear controllers, including Linear Quadratic Regulator (LQR) for AGV position tracking and a mixed H2/H∞ and Proportional Derivative (PD) controller for drone attitude and position control. The results demonstrate significant positional errors under packet loss, emphasizing the necessity for predictive methods to mitigate these effects. To tackle this problem, the thesis proposes the use of Long Short-Term Memory (LSTM) networks, which excel at retaining long-term dependencies, for predicting missing data during packet loss. A comparative analysis involving LSTM, Gated Recurrent Units (GRUs), Linear Interpolation Predictor (LIP), and the Memory Consensus Protocol (MCP) highlights LSTM’s superior accuracy in trajectory prediction for AGVs and drones. The LSTM model was further integrated into closed-loop systems, where it significantly enhanced performance under feedback dynamics and external disturbances, outperforming MCP and LIP methods. The robustness of the proposed method was validated through extensive testing under realistic scenarios, including curvy trajectories, such as circular and figure-eight paths for AGVs and spiral trajectories and figure-eight paths for drones, as well as scenarios with initial offset conditions, actuator saturation, and the presence of external disturbances like sensor noise and wind. These tests demonstrated LSTM’s capability to maintain formation control accuracy in challenging environments, making it well-suited for practical applications in AGVs and drones. This thesis also compares the LSTM-based linear control strategies with highly nonlinear controllers, such as back-stepping control and event-based sliding mode control. The comparative evaluation includes challenging conditions, such as time-varying delays, which initially caused oscillations in the system. By applying an input-shaping filter, the oscillatory behavior is mitigated, and the stability of the system is restored. The LSTM-based linear control strategy exhibits performance that is competitive with nonlinear controllers, showcasing its effectiveness as a robust alternative for handling formation control challenges in multi-robot systems. These results demonstrate that combining LSTM with linear controllers enhances their performance under uncertainties such as packet loss, improving their robustness to a level comparable with nonlinear controllers.","abstract_html":"This thesis addresses the challenges posed by communication packet loss in leader-follower formation control of multi-robot systems, with a focus on Automated Guided Vehicles (AGVs) and drones. It presents a comprehensive analysis of how packet loss affects linear controllers, including Linear Quadratic Regulator (LQR) for AGV position tracking and a mixed H2/H∞ and Proportional Derivative (PD) controller for drone attitude and position control. The results demonstrate significant positional errors under packet loss, emphasizing the necessity for predictive methods to mitigate these effects. To tackle this problem, the thesis proposes the use of Long Short-Term Memory (LSTM) networks, which excel at retaining long-term dependencies, for predicting missing data during packet loss. A comparative analysis involving LSTM, Gated Recurrent Units (GRUs), Linear Interpolation Predictor (LIP), and the Memory Consensus Protocol (MCP) highlights LSTM’s superior accuracy in trajectory prediction for AGVs and drones. The LSTM model was further integrated into closed-loop systems, where it significantly enhanced performance under feedback dynamics and external disturbances, outperforming MCP and LIP methods. The robustness of the proposed method was validated through extensive testing under realistic scenarios, including curvy trajectories, such as circular and figure-eight paths for AGVs and spiral trajectories and figure-eight paths for drones, as well as scenarios with initial offset conditions, actuator saturation, and the presence of external disturbances like sensor noise and wind. These tests demonstrated LSTM’s capability to maintain formation control accuracy in challenging environments, making it well-suited for practical applications in AGVs and drones. This thesis also compares the LSTM-based linear control strategies with highly nonlinear controllers, such as back-stepping control and event-based sliding mode control. The comparative evaluation includes challenging conditions, such as time-varying delays, which initially caused oscillations in the system. By applying an input-shaping filter, the oscillatory behavior is mitigated, and the stability of the system is restored. The LSTM-based linear control strategy exhibits performance that is competitive with nonlinear controllers, showcasing its effectiveness as a robust alternative for handling formation control challenges in multi-robot systems. These results demonstrate that combining LSTM with linear controllers enhances their performance under uncertainties such as packet loss, improving their robustness to a level comparable with nonlinear controllers.","abstract_has_math":false,"creators":["Sedghi, Leila"],"institution":"University College Cork","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Pesch, Dirk H J","Brown, Kenneth"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025","date_published":"2025","updated_at":"2026-07-24T01:46:44Z","subjects":["Formation control","Automated Guided Vehicles (AGVs)","Drones","Long Short-Term Memory (LSTM)","Prediction","Packet loss formation control"],"languages":["en"],"rights":["© 2025, Leila Sedghi."],"rights_urls":["https://creativecommons.org/licenses/by/4.0/"],"identifier_entries":[]},"links":{"outbound_url":"https://hdl.handle.net/10468/18509","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Pesch, Dirk H J","Brown, Kenneth"]},{"key":"dc:creator","label":"Author","values":["Sedghi, Leila"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2026-02-09T08:52:06Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2026-02-09T08:52:06Z"]},{"key":"dc:date.issued","label":"Date","values":["2025"]},{"key":"dc:publisher","label":"Institution","values":["University College Cork"]},{"key":"dc:type","label":"Dc Type","values":["Doctoral thesis"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Doctoral"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["PhD - Doctor of Philosophy"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Formation control","Automated Guided Vehicles (AGVs)","Drones","Long Short-Term Memory (LSTM)","Prediction","Packet loss formation control"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["© 2025, Leila Sedghi."]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://creativecommons.org/licenses/by/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/10468/18509"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["This thesis addresses the challenges posed by communication packet loss in leader-follower formation control of multi-robot systems, with a focus on Automated Guided Vehicles (AGVs) and drones. It presents a comprehensive analysis of how packet loss affects linear controllers, including Linear Quadratic Regulator (LQR) for AGV position tracking and a mixed H2/H∞ and Proportional Derivative (PD) controller for drone attitude and position control. The results demonstrate significant positional errors under packet loss, emphasizing the necessity for predictive methods to mitigate these effects. To tackle this problem, the thesis proposes the use of Long Short-Term Memory (LSTM) networks, which excel at retaining long-term dependencies, for predicting missing data during packet loss. A comparative analysis involving LSTM, Gated Recurrent Units (GRUs), Linear Interpolation Predictor (LIP), and the Memory Consensus Protocol (MCP) highlights LSTM’s superior accuracy in trajectory prediction for AGVs and drones. The LSTM model was further integrated into closed-loop systems, where it significantly enhanced performance under feedback dynamics and external disturbances, outperforming MCP and LIP methods. The robustness of the proposed method was validated through extensive testing under realistic scenarios, including curvy trajectories, such as circular and figure-eight paths for AGVs and spiral trajectories and figure-eight paths for drones, as well as scenarios with initial offset conditions, actuator saturation, and the presence of external disturbances like sensor noise and wind. These tests demonstrated LSTM’s capability to maintain formation control accuracy in challenging environments, making it well-suited for practical applications in AGVs and drones. This thesis also compares the LSTM-based linear control strategies with highly nonlinear controllers, such as back-stepping control and event-based sliding mode control. The comparative evaluation includes challenging conditions, such as time-varying delays, which initially caused oscillations in the system. By applying an input-shaping filter, the oscillatory behavior is mitigated, and the stability of the system is restored. The LSTM-based linear control strategy exhibits performance that is competitive with nonlinear controllers, showcasing its effectiveness as a robust alternative for handling formation control challenges in multi-robot systems. These results demonstrate that combining LSTM with linear controllers enhances their performance under uncertainties such as packet loss, improving their robustness to a level comparable with nonlinear controllers."]},{"key":"dc:format.mimetype","label":"Dc Format Mimetype","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Formation control of multiple robots under packet loss"]}]}],"canonical_facts":{"dc:contributor.advisor":["Pesch, Dirk H J","Brown, Kenneth"],"dc:creator":["Sedghi, Leila"],"dc:date.accessioned":["2026-02-09T08:52:06Z"],"dc:date.available":["2026-02-09T08:52:06Z"],"dc:date.issued":["2025"],"dc:description.abstract":["This thesis addresses the challenges posed by communication packet loss in leader-follower formation control of multi-robot systems, with a focus on Automated Guided Vehicles (AGVs) and drones. It presents a comprehensive analysis of how packet loss affects linear controllers, including Linear Quadratic Regulator (LQR) for AGV position tracking and a mixed H2/H∞ and Proportional Derivative (PD) controller for drone attitude and position control. The results demonstrate significant positional errors under packet loss, emphasizing the necessity for predictive methods to mitigate these effects. To tackle this problem, the thesis proposes the use of Long Short-Term Memory (LSTM) networks, which excel at retaining long-term dependencies, for predicting missing data during packet loss. A comparative analysis involving LSTM, Gated Recurrent Units (GRUs), Linear Interpolation Predictor (LIP), and the Memory Consensus Protocol (MCP) highlights LSTM’s superior accuracy in trajectory prediction for AGVs and drones. The LSTM model was further integrated into closed-loop systems, where it significantly enhanced performance under feedback dynamics and external disturbances, outperforming MCP and LIP methods. The robustness of the proposed method was validated through extensive testing under realistic scenarios, including curvy trajectories, such as circular and figure-eight paths for AGVs and spiral trajectories and figure-eight paths for drones, as well as scenarios with initial offset conditions, actuator saturation, and the presence of external disturbances like sensor noise and wind. These tests demonstrated LSTM’s capability to maintain formation control accuracy in challenging environments, making it well-suited for practical applications in AGVs and drones. This thesis also compares the LSTM-based linear control strategies with highly nonlinear controllers, such as back-stepping control and event-based sliding mode control. The comparative evaluation includes challenging conditions, such as time-varying delays, which initially caused oscillations in the system. By applying an input-shaping filter, the oscillatory behavior is mitigated, and the stability of the system is restored. The LSTM-based linear control strategy exhibits performance that is competitive with nonlinear controllers, showcasing its effectiveness as a robust alternative for handling formation control challenges in multi-robot systems. These results demonstrate that combining LSTM with linear controllers enhances their performance under uncertainties such as packet loss, improving their robustness to a level comparable with nonlinear controllers."],"dc:format.mimetype":["application/pdf"],"dc:identifier.uri":["https://hdl.handle.net/10468/18509"],"dc:language.iso":["en"],"dc:publisher":["University College Cork"],"dc:rights":["© 2025, Leila Sedghi."],"dc:rights.uri":["https://creativecommons.org/licenses/by/4.0/"],"dc:subject":["Formation control","Automated Guided Vehicles (AGVs)","Drones","Long Short-Term Memory (LSTM)","Prediction","Packet loss formation control"],"dc:title":["Formation control of multiple robots under packet loss"],"dc:type":["Doctoral thesis"],"dc:type.qualificationlevel":["Doctoral"],"dc:type.qualificationname":["PhD - Doctor of Philosophy"]},"updated_at":"2026-07-24T01:46:44Z"}