cs.ROApr 15, 2026

Scale-Invariant Sampling in Multi-Arm Bandit Motion Planning for Object Extraction

Authors: Servet B. Bayraktar, Andreas Orthey, Marc Toussaint

Organizations: Technical University of Berlin

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Object extraction tasks often occur in disassembly problems, where bolts, screws, or pins have to be removed from tight, narrow spaces. In such problems, the distance to the environment is often on the millimeter scale. Sampling-based planners can solve such problems and provide completeness guarantees. However, sampling becomes a bottleneck, since almost all motions will result in collisions with the environment. To overcome this problem, we propose a novel scale-invariant sampling strategy which explores the configuration space using a grow-shrink search to find useful, high-entropy sampling scales. Once a useful sampling scale has been found, our framework exploits this scale by using a principal components analysis (PCA) to find useful directions for object extraction. We embed this sampler into a multi-arm bandit rapidly-exploring random tree (MAB-RRT) planner and test it on eight challenging 3D object extraction scenarios, involving bolts, gears, rods, pins, and sockets. To evaluate our framework, we compare it with classical sampling strategies like uniform sampling, obstacle-based sampling, and narrow-passage sampling, and with modern strategies like mate vectors, physics-based planning, and disassembly breadth first search. Our experiments show that scale-invariant sampling improves success rate by one order of magnitude on 7 out of 8 scenarios. This demonstrates that scale-invariant sampling is an important concept for general purpose object extraction in disassembly tasks.

Figures & tables

Figure 1: Overview about scale-invariant sampling in MAB-RRT. As input, we use a start configuration xstartx_{\text{start}} and a goal region XgoalX_{\text{goal}} . We then search for a high-information entropy scale and initialize the multi-arm bandit with this scale. We then start a MAB-RRT loop, where an arm is chose, the chosen sampler is used for one RRT update step, and the arm reward is updated. This loop continues until a termination condition is true or a path π\pi has been found which solves the problem.
Figure 2: Visualization of scale-invariant sampling in tunnel passages of different sizes (5 units, 10 units, and 15 units). Final burn-in radius is shown as , and path are shown depending on the type of active sampler: uniform, positive principal component, and negative principal component. Start is marked with , goal with . Valid samples are shown as uniform, positive principal component, negative principal component, and burn-in valid samples as .
Figure 3: Environment Comparison: Start vs. End states for selected environments. Each pair shows the initial (left) and final (right) configuration.
Figure 4: Comparison of success rate using planners MAB-RRT (ours), RRT + Bridge Sampling [ 24 ] , RRT + Gaussian Sampling [ 8 ] , RRT + Obstacle Sampling [ 2 ] , MateVec-TRRT [ 17 ] , BK-RRT [ 55 , 52 ] , BFS [ 52 ] ,
Figure 5: Final sampling results across environments (radius set to 1e-6 vs. 25.0). Final radius from grow-shrink shown as , initial radius as . Tree edges shown as . Valid samples: uniform, positive principal component, negative principal component, scale-invariant sampler. Start marked with , goal with .
Figure 6: T-Bolt Experiment: MAB dynamics during planning. (a) UCB scores adjust dynamically as the planner gathers reward information. (b) Rapid reward accumulation for cylinder arms indicates their effectiveness in generating valid samples within the narrow passage, while the uniform arm’s reward surges once free space is reached. Uniform, PCA positive, PCA negative.
More in this topic

Sampling-Based Motion Planning

Sampling-based motion planning uses random or guided state and control samples to find feasible robot paths through complex spaces.

See all 96 papers

Explore similar work

CardsList
Aug 9, 2026

PEEL: Parallel Extraction for Long-Horizon Disassembly Planning via Scale-Invariant Sampling

Long-horizon multi-part object disassembly requires robots to compute feasible sequences of collision-free removal motions, even in the presence of tight, narrow escape corridors. To efficiently solve such disassembly problems, we propose Parallel Extraction for Long-Horizon Disassembly (PEEL), an algorithm which efficiently computes disassembly motions for object assemblies and feeds them to a robot manipulator for execution. PEEL uses sampling-based motion planning to compute single-object motions through the use of a scale-invariant sampling scheme, where the object scale is estimated in a burn-in phase and a subsequent directional sampler exploits the scale. This sampling scheme is integrated into a multi-arm bandit rapidly-exploring random tree (MAB-RRT) planner, which switches between different samplers depending on the reward signal received. Using MAB-RRT, the PEEL algorithm runs a batch of planners in parallel to obtain an ordered graph specifying the sequence in which object parts have to be removed. We show that MAB-RRT can efficiently solve single-part disassemblies with 100 percent success rate on 76 assemblies, and that it is robust to its parameters. By integrating MAB-RRT into PEEL, we solve four long-horizon disassembly problems using the Fetch manipulator robot involving 10 to 17 individual object parts.

Servet B. Bayraktar, Andreas Orthey, Zachary Kingston +1

Long-Horizon Robotic ManipulationRobotic ManipulationSampling-Based Motion Planning
AskPDF
May 6, 2024

Greedy Heuristics for Sampling-Based Motion Planning in High-Dimensional State Spaces

Informed sampling techniques accelerate the convergence of sampling-based motion planners by biasing sampling toward regions of the state space that are most likely to yield better solutions. However, when the current solution path contains redundant or tortuous segments, the resulting informed subset may remain unnecessarily large, slowing convergence. Our prior work addressed this issue by introducing the greedy informed set, which reduces the sampling region based on the maximum heuristic cost along the current solution path. In this article, we formally characterize the behavior of the greedy informed set within Rapidly-exploring Random Tree (RRT*)-like planners and analyze how greedy sampling affects exploration and asymptotic optimality. We then present Greedy RRT* (G-RRT*), a bi-directional anytime variant of RRT* that leverages the greedy informed set to focus sampling in the most promising regions of the search space. Experiments on abstract planning benchmarks, manipulation tasks from the MotionBenchMaker dataset, and a dual-arm Barrett WAM problem demonstrate that G-RRT* rapidly finds initial solutions and converges asymptotically to optimal paths, outperforming state-of-the-art sampling-based planners.

Phone Thiha Kyaw, Anh Vu Le, Rajesh Elara Mohan +1

Space & Terrestrial Autonomous Robotic Systems (STARS) Laboratory, University of Toronto Institute for Aerospace Studies, 4925 Dufferin Sreet, Toronto, M3H 5T6, Ontario, Canada. · Advanced Intelligent Technology Research Group, Faculty of Electrical and Electronics Engineering, Ton Duc Thang University, Ho Chi Minh City, 700000, Vietnam. · ROAR Lab, Engineering Product Development, Singapore University of Technology and Design, Singapore, 487372, Singapore.

Sampling-Based Motion PlanningTree Search
AskPDF
Mar 6, 2026

Stability-Guided Exploration for Diverse Motion Generation

Scaling up datasets is highly effective in improving the performance of deep learning models, including in the field of robot learning. However, data collection still proves to be a bottleneck. Approaches relying on collecting human demonstrations are labor-intensive and inherently limited: they tend to be narrow, task-specific, and fail to adequately explore the full space of feasible states. Synthetic data generation could remedy this, but current techniques mostly rely on local trajectory optimization and fail to find diverse solutions. In this work, we propose a novel method capable of finding diverse long-horizon manipulations through black-box simulation. We achieve this by combining an RRT-style search with sampling-based MPC, together with a novel sampling scheme that guides the exploration toward stable configurations. Specifically, we sample from a manifold of stable states while growing a search tree directly through simulation, without restricting the planner to purely stable motions. We demonstrate the method's ability to discover diverse manipulation strategies, including pushing, grasping, pivoting, throwing, and tool use, across different robot morphologies, without task-specific guidance.

Eckart Cobo-Briesewitz, Tilman Burghoff, Denis Shcherba +2

Technische Universität Berlin · LAAS-CNRS · Robotics Institute Germany (RIG)

Robot Motion GenerationLong-Horizon Robotic ManipulationMotion PlanningSynthetic Data Generation for RoboticsSampling-Based Motion Planning
AskPDF