FlexWorm: Primitive-augmented Hybrid Contact-motion Planning for Suction-based Multi-segment Deformable Robots
Paper Guide Brief
Reading Brief
The paper presents FlexWorm, a planning framework for multi-segment suction-based soft robots that combines block-wise inverse kinematics hybrid search (IKHS) with a primitive-augmented hybrid search (PaHS) using learned embeddings for efficient motion planning on complex 3D surfaces.
Central Claim
A hierarchical hybrid planner (IKHS+PaHS) that integrates discrete adhesion switching and continuous body deformation for suction-based soft robots, using block-wise IK to reduce search complexity and a learned primitive library to accelerate planning.
Contribution
A hierarchical hybrid planner (IKHS+PaHS) that integrates discrete adhesion switching and continuous body deformation for suction-based soft robots, using block-wise IK to reduce search complexity and a learned primitive library to accelerate planning.
Why It Matters
This contribution matters because it enables autonomous long-horizon navigation for suction-based soft robots without hand-designed gaits, achieving order-of-magnitude speedups while maintaining feasibility through hybrid search and learned primitive reuse.
Prerequisites
block-wise IK hybrid search, primitive-augmented hybrid search, learned observation-primitive embedding, best-first search, inverse kinematics
Atlas Placement
Robot Motion Planning (subfield)
Read If
You care about block-wise IK hybrid search, primitive-augmented hybrid search, learned observation-primitive embedding.
Skip If
You only care about planning success rate, planning time.
Noosaga Placements
- The paper focuses on motion planning for soft robots, introducing IKHS and PaHS for contact-motion planning.This work presents a planning framework for serial multi-segment soft robots...Its core, block-wise IK hybrid search (IKHS), performs best-first search over feasible adhesion transitions...
- Search-Based Planningframework90%IKHS is a best-first search over hybrid states, which is a form of search-based planning.IKHS performs best-first search over hybrid states...
- Continuum and Soft Robot Modelingframework85%The paper models soft segments using screw theory and quasi-static load-deformation feasibility, which falls under continuum and soft robot modeling.Segment kinematics via screw theory: Let Ti P SEp3q be the homogeneous transform of link i.For a segment deformation to be physically realizable, a gravity-conditioned quasi-static feasibility constraint is enforced...
- The method heavily relies on screw-theory-based kinematics and quasi-static load-deformation feasibility for soft segments.Segment kinematics via screw theory: Let Ti P SEp3q be the homogeneous transform of link i.For a segment deformation to be physically realizable, a gravity-conditioned quasi-static feasibility constraint is enforced...
- Learning-Based Motion Planningframework80%PaHS uses a learned embedding to retrieve motion primitives, which is a learning-based approach to motion planning.primitive-augmented hybrid search (PaHS) uses a learned observation–primitive embedding to retrieve short validated motion segments...
- PaHS uses a learned observation-primitive embedding for retrieval, which is a learning-based component.primitive-augmented hybrid search (PaHS) uses a learned observation–primitive embedding to retrieve short validated motion segments...
- The core IKHS is a best-first search over hybrid states, and the guide path uses A* search.IKHS performs best-first search over hybrid states...The path is generated by A‹ search with the edge cost defined below...
- Data-Driven and Learning-Based Manipulationframework50%The paper addresses suction-based adhesion for locomotion, which is a form of manipulation of the environment, but the focus is on planning rather than manipulation.Multi-segment suction-based soft robots combine geometric adaptability, compliance, and safe contact...
Abstract
Multi-segment suction-based soft robots are promising for inspection and maintenance in confined or fragile environments, but existing approaches still depend heavily on manually designed gaits and environment-specific motion scripts. This work presents a planning framework for serial multi-segment soft robots with deformable body segments and boundary suction pads. The formulation targets full 3D navigation on complex surfaces and explicitly handles discrete adhesion switching and continuous body deformation under geometric, collision, and quasi-static feasibility constraints, while remaining agnostic to the specific actuation realization used to produce segment deformation. Its core, block-wise IK hybrid search (IKHS), performs best-first search over feasible adhesion transitions while solving inverse kinematics only on induced free blocks. On top of IKHS, primitive-augmented hybrid search (PaHS) uses a learned observation--primitive embedding to retrieve short validated motion segments for fast local proposal, with fallback to standard IKHS branching when retrieval fails. In simulation, the framework consistently outperforms controlled baselines in planning success, transition quality, and efficiency across diverse terrains. PaHS matches IKHS in success rate while substantially reducing planning time. Repeated hardware experiments on a pneumatic multi-segment soft robot further demonstrate executability and online recovery under actuation and adhesion uncertainty.
Paper Context
Classified from the full extracted paper text (40,074 characters). The Paper Guide brief above is the user-facing synthesis; raw context is kept out of the page.
Full-paper context sent 40,074 of 40,074 extracted characters to classification.