Toward Generalist Neural Motion Planners for Robotic Manipulators: Challenges and Opportunities

Abstract

State-of-the-art generalist manipulation policies have enabled the deployment of robotic manipulators in unstructured human environments. However, these frameworks struggle in cluttered environments primarily because they utilize auxiliary modules for low-level motion planning and control. Motion planning remains challenging due to the high dimensionality of the robot's configuration space and the presence of workspace obstacles. Neural motion planners have enhanced motion planning efficiency by offering fast inference and effectively handling the inherent multi-modality of the motion planning problem. Despite such benefits, current neural motion planners often struggle to generalize to unseen, out-of-distribution planning settings. This paper reviews and analyzes the state-of-the-art neural motion planners, highlighting both their benefits and limitations. It also outlines a path toward establishing generalist neural motion planners capable of handling domain-specific challenges. For a list of the reviewed papers, please refer to https://davoodsz.github.io/planning-manip-survey.github.io/.

References in corpus (30)

• Control Barrier Function Based Quadratic Programs for Safety Critical Systems
• Normalizing Flows: An Introduction and Review of Current Methods
• Informed RRT*: Optimal Sampling-based Path Planning Focused via Direct Sampling of an Admissible Ellipsoidal Heuristic
• Deep Neural Networks and Tabular Data: A Survey
• Voxblox: Incremental 3D Euclidean Signed Distance Fields for On-Board MAV Planning
• Highway Networks
• Potential Functions based Sampling Heuristic For Optimal Path Planning
• Continuous-Time Gaussian Process Motion Planning via Probabilistic Inference
• Batch Informed Trees (BIT*): Informed Asymptotically Optimal Anytime Search
• Adaptively Informed Trees (AIT*): Fast Asymptotically Optimal Path Planning through Adaptive Heuristics
• Shortest Paths in Graphs of Convex Sets
• Multimodal Trajectory Optimization for Motion Planning
• MotionBenchMaker: A Tool to Generate and Benchmark Motion Planning Datasets
• Language Models as Zero-Shot Trajectory Generators
• Motion Planning for Robotics: A Review for Sampling-based Planners
• An Asymptotically-Optimal Sampling-Based Algorithm for Bi-directional Motion Planning
• Automating Surgical Peg Transfer: Calibration with Deep Learning Can Exceed Speed, Accuracy, and Consistency of Humans
• Safety and Liveness Guarantees through Reach-Avoid Reinforcement Learning
• Neural Manipulation Planning on Constraint Manifolds
• NeSLAM: Neural Implicit Mapping and Self-Supervised Feature Tracking With Depth Completion and Denoising
• Learning Implicit Priors for Motion Optimization
• Configuration Space Decomposition for Scalable Proxy Collision Checking in Robot Planning and Control
• End-to-end deep learning-based framework for path planning and collision checking: bin picking application
• Reaching Through Latent Space: From Joint Statistics to Path Planning in Manipulation
• A Survey on the Integration of Machine Learning with Sampling-based Motion Planning
• A Gaussian variational inference approach to motion planning
• GraphDistNet: A Graph-based Collision-distance Estimator for Gradient-based Trajectory Optimization
• Approximating Constraint Manifolds Using Generative Models for Sampling-Based Constrained Motion Planning
• Transformer-Enhanced Motion Planner: Attention-Guided Sampling for State-Specific Decision Making
• Surgical Neural Radiance Fields from One Image