LSTM-Enhanced Deep Reinforcement Learning for Robust Trajectory Tracking Control of Skid-Steer Mobile Robots Under Terra-Mechanical Constraints

Simple item page
dc.article.number74
dc.catalogadorgjm
dc.contributor.authorAlcayaga, José Manuel
dc.contributor.authorMenéndez, Oswaldo Anibal
dc.contributor.authorTorres Torriti, Miguel Attilio
dc.contributor.authorVásconez, Juan Pablo
dc.contributor.authorArévalo Ramírez, Tito
dc.contributor.authorPrado Romo, Álvaro Javier
dc.date.accessioned2025-06-04T15:58:51Z
dc.date.available2025-06-04T15:58:51Z
dc.date.issued2025
dc.description.abstractAutonomous navigation in mining environments is challenged by complex wheel–terrain interaction, traction losses caused by slip dynamics, and sensor limitations. This paper investigates the effectiveness of Deep Reinforcement Learning (DRL) techniques for the trajectory tracking control of skid-steer mobile robots operating under terra-mechanical constraints. Four state-of-the-art DRL algorithms, i.e., Proximal Policy Optimization (PPO), Deep Deterministic Policy Gradient (DDPG), Twin Delayed DDPG (TD3), and Soft Actor–Critic (SAC), are selected to evaluate their ability to generate stable and adaptive control policies under varying environmental conditions. To address the inherent partial observability in real-world navigation, this study presents an original approach that integrates Long Short-Term Memory (LSTM) networks into DRL-based controllers. This allows control agents to retain and leverage temporal dependencies to infer unobservable system states. The developed agents were trained and tested in simulations and then assessed in field experiments under uneven terrain and dynamic model parameter changes that lead to traction losses in mining environments, targeting various trajectory tracking tasks, including lemniscate and squared-type reference trajectories. This contribution strengthens the robustness and adaptability of DRL agents by enabling better generalization of learned policies compared with their baseline counterparts, while also significantly improving trajectory tracking performance. In particular, LSTM-based controllers achieved reductions in tracking errors of 10%, 74%, 21%, and 37% for DDPG-LSTM, PPO-LSTM, TD3-LSTM, and SAC-LSTM, respectively, compared with their non-recurrent counterparts. Furthermore, DDPG-LSTM and TD3-LSTM reduced their control effort through the total variation in control input by 15% and 20% compared with their respective baseline controllers, respectively. Findings from this work provide valuable insights into the role of memory-augmented reinforcement learning for robust motion control in unstructured and high-uncertainty environments.
dc.fechaingreso.objetodigital2025-06-04
dc.format.extent37 páginas
dc.fuente.origenORCID
dc.identifier.doi10.3390/robotics14060074
dc.identifier.urihttps://doi.org/10.3390/robotics14060074
dc.identifier.urihttps://repositorio.uc.cl/handle/11534/104577
dc.information.autorucEscuela de Ingeniería; Torres Torriti, Miguel Attilio; 0000-0002-7904-7981; 96590
dc.information.autorucEscuela de Ingeniería; Arévalo Ramírez, Tito; 0000-0003-2542-6545; 1300544
dc.issue.numero6
dc.language.isoen
dc.nota.accesocontenido completo
dc.revistaRobotics
dc.rightsacceso abierto
dc.rights.licenseCC BY 4.0 Attribution 4.0 International
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectDeep reinforcement learning
dc.subjectTrajectory tracking
dc.subjectRecurrent neural network
dc.subjectTerra-mechanical constraints
dc.subject.ddc600
dc.subject.deweyTecnologíaes_ES
dc.subject.ods09 Industry, innovation and infrastructure
dc.subject.odspa09 Industria, innovación e infraestructura
dc.titleLSTM-Enhanced Deep Reinforcement Learning for Robust Trajectory Tracking Control of Skid-Steer Mobile Robots Under Terra-Mechanical Constraints
dc.typeartículo
dc.volumen14
sipa.codpersvinculados96590
sipa.codpersvinculados1300544
sipa.trazabilidadORCID;2025-06-03
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
robotics-14-00074.pdf
Size:
13.54 MB
Format:
Adobe Portable Document Format
Description:
Download
Collections
Artículos de revistas