{"paper":{"title":"3D RL-DWA: A Hybrid Reinforcement Learning and Dynamic Window Approach for Goal-Directed Local Navigation in Multi-DoF Robots","license":"http://creativecommons.org/licenses/by/4.0/","headline":"A hybrid RL and DWA controller improves deformation and path completion for microrobots in 3D constrained spaces.","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Chiara Castellani, Domenico Prattichizzo, Enrico Turco","submitted_at":"2026-05-12T19:37:32Z","abstract_excerpt":"In this paper, we present a novel hybrid approach that combines Reinforcement Learning (RL) with Dynamic Window Approach (DWA) for adaptive 3D local navigation of high-degree-of-freedom robotic systems. Our method leverages sparse point cloud data to dynamically adjust both the motion and the shape of a deformable microrobot, enabling the system to navigate toward a goal in complex, constrained environments while maximizing the occupied volume. We evaluate our framework in a simulated vascular network. Experimental results, based on 1080 trials, indicate that integrating RL with a DWA-based lo"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"integrating RL with a DWA-based local planner significantly enhances both deformation and navigation capabilities compared to a pure RL and a model-based methods. In particular, the proposed autonomous controller consistently achieves high deformation and near-perfect path completion during training and maintains robust performance in unseen scenarios.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The simulated vascular network and sparse point-cloud sensor model are sufficiently representative of real-world dynamics, contact forces, and sensing noise for the performance gains to transfer.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A hybrid RL-DWA controller achieves high deformation and near-perfect path completion for deformable microrobots navigating simulated 3D vascular networks from sparse point clouds.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A hybrid RL and DWA controller improves deformation and path completion for microrobots in 3D constrained spaces.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"e173d34c18c1cd8e447f5a75d2630ba7ec01019379514961b29ccbc4d8b0ee73"},"source":{"id":"2605.12689","kind":"arxiv","version":1},"verdict":{"id":"819edd84-5447-4ed7-b877-f04b0de4362f","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T20:17:07.159888Z","strongest_claim":"integrating RL with a DWA-based local planner significantly enhances both deformation and navigation capabilities compared to a pure RL and a model-based methods. In particular, the proposed autonomous controller consistently achieves high deformation and near-perfect path completion during training and maintains robust performance in unseen scenarios.","one_line_summary":"A hybrid RL-DWA controller achieves high deformation and near-perfect path completion for deformable microrobots navigating simulated 3D vascular networks from sparse point clouds.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The simulated vascular network and sparse point-cloud sensor model are sufficiently representative of real-world dynamics, contact forces, and sensing noise for the performance gains to transfer.","pith_extraction_headline":"A hybrid RL and DWA controller improves deformation and path completion for microrobots in 3D constrained spaces."},"references":{"count":31,"sample":[{"doi":"","year":2023,"title":"Path planning techniques for mobile robots: Review and prospect,","work_id":"6525e313-f647-494b-913b-642bb69fee4b","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2007,"title":"C. Laugier and R. Chatila,Autonomous navigation in dynamic environ- ments. Springer, 2007, vol. 35","work_id":"a29c42bb-6ca2-41f1-a310-38892c2fb7b5","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"Fast motion planning for high-dof robot systems using hierarchical system identification,","work_id":"214c3dfe-ec88-438a-8288-efe044f98fda","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"A rrt based path planning scheme for multi-dof robots in unstructured environments,","work_id":"3dddd54b-ee24-4ab3-95fa-169bd34005bb","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1997,"title":"The dynamic window approach to collision avoidance","work_id":"da13a512-795a-4556-8521-692996fedaf7","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":31,"snapshot_sha256":"04aecf947bfdd2fa1aca22dc7c05c81f493a449805c5aa4f816246246fb5b050","internal_anchors":1},"formal_canon":{"evidence_count":1,"snapshot_sha256":"cfef44138d47fff908eeeb3fea8a8f490e0753d943dd44b3d951c21fd17e7b73"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}