Comparative benchmark of sampling-based and DRL motion planning methods for industrial robotic arms

Vista sencilla de ítem
dc.contributor.authorFidalgo Astorquia, Ignacio
dc.contributor.authorVillate Castillo, Guillermo
dc.contributor.authorTellaeche Iglesias, Alberto
dc.contributor.authorVazquez, Juan-Ignacio
dc.date.accessioned2025-10-02T09:00:51Z
dc.date.available2025-10-02T09:00:51Z
dc.date.issued2025-08-25
dc.date.updated2025-10-02T09:00:51Z
dc.description.abstractThis study presents a comprehensive comparison between classical sampling-based motion planners from the Open Motion Planning Library (OMPL) and a learning-based planner based on Soft Actor–Critic (SAC) for motion planning in industrial robotic arms. Using a UR3e robot equipped with an RG2 gripper, we constructed a large-scale dataset of over 100,000 collision-free trajectories generated with MoveIt-integrated OMPL planners. These trajectories were used to train a DRL agent via curriculum learning and expert demonstrations. Both approaches were evaluated on key metrics such as planning time, success rate, and trajectory smoothness. Results show that the DRL-based planner achieves higher success rates and significantly lower planning times, producing more compact and deterministic trajectories. Time-optimal parameterization using TOPPRA ensured the dynamic feasibility of all trajectories. While classical planners retain advantages in zero-shot adaptability and environmental generality, our findings highlight the potential of DRL for real-time and high-throughput motion planning in industrial contexts. This work provides practical insights into the trade-offs between traditional and learning-based planning paradigms, paving the way for hybrid architectures that combine their strengths.en
dc.description.sponsorshipThis research was carried out within the project ACROBA, which has received funding from the European Union’s Horizon 2020 research and innovation programme, under grant agreement No. 101017284en
dc.identifier.citationAstorquia, I. F., Villate-Castillo, G., Tellaeche, A., & Vazquez, J.-I. (2025). Comparative benchmark of sampling-based and DRL motion planning methods for industrial robotic arms. Sensors, 25(17). https://doi.org/10.3390/S25175282
dc.identifier.doi10.3390/S25175282
dc.identifier.eissn1424-8220
dc.identifier.urihttps://hdl.handle.net/20.500.14454/3825
dc.language.isoeng
dc.publisherMultidisciplinary Digital Publishing Institute (MDPI)
dc.rights© 2025 by the authors
dc.subject.otherCurriculum learning
dc.subject.otherDeep reinforcement learning (DRL)
dc.subject.otherHybrid motion planning
dc.subject.otherIndustrial robotics
dc.subject.otherMotion planning
dc.subject.otherOpen Motion Planning Library (OMPL)
dc.subject.otherSampling-based planners
dc.titleComparative benchmark of sampling-based and DRL motion planning methods for industrial robotic armsen
dc.typejournal article
dcterms.accessRightsopen access
oaire.citation.issue17
oaire.citation.titleSensors
oaire.citation.volume25
oaire.licenseConditionhttps://creativecommons.org/licenses/by/4.0/
oaire.versionVoR
Archivos
Bloque original
Mostrando 1 - 1 de 1
Miniatura
Nombre:
fidalgo_comparative_2025.pdf
Tamaño:
11.83 MB
Formato:
Adobe Portable Document Format
Descargar
Colecciones
Artículos