Bayesian Local Sampling-Based Planning

@article{Lai2020BayesianLS,
  title={Bayesian Local Sampling-Based Planning},
  author={Tin Lai and Philippe Morere and Fabio Ramos and Gilad Francis},
  journal={IEEE Robotics and Automation Letters},
  year={2020},
  volume={5},
  pages={1954-1961}
}
Sampling-based planning is the predominant paradigm for motion planning in robotics. Most sampling-based planners use a global random sampling scheme to guarantee probabilistic completeness. However, most schemes are often inefficient as the samples drawn from the global proposal distribution, and do not exploit relevant local structures. Local sampling-based motion planners, on the other hand, take sequential decisions of random walks to samples valid trajectories in configuration space… 

Figures and Tables from this paper

Adaptively Exploits Local Structure With Generalised Multi-Trees Motion Planning
TLDR
This letter proposes Rapidly-exploring Random Forest (rrf*)—a generalised multi-trees motion planner that combines the rapid exploring property of tree-based methods and adaptively learns to deploys a Bayesian local sampling strategy in regions that are deemed to be bottlenecks.
Parallelised Diffeomorphic Sampling-based Motion Planning
TLDR
PDMP is a novel parallelised framework that uses bijective and differentiable mappings, or diffeomorphisms, to transform sampling distributions of sampling-based motion planners, in a manner akin to normalising flows.
Rapid Replanning in Consecutive Pick-and-Place Tasks with Lazy Experience Graph
TLDR
This work proposes Lazy Tree-based Replanner (LT R*)—a novel hybrid planner that inherits the rapid planning nature of existing anytime incremental sampling-based planners, and at the same time allows subsequent tasks to leverage prior experience via a lazy experience graph.
Rapidly-exploring Random Forest: Adaptively Exploits Local Structure with Generalised Multi-Trees Motion Planning
TLDR
This paper proposes Rapidly-exploring Random Forest (R R F*)—a generalised multitrees motion planner that combines the rapid exploring property of tree-based methods and adaptively learns to deploys a Bayesian local sampling strategy in regions that are deemed to be bottlenecks.
sbp-env: A Python Package for Sampling-based Motion Planner and Samplers
TLDR
Sampling-based motion planners are robust methods that avoid explicitly constructing the often intractable high-dimensional configuration space (C-Space).
sbp-env: A Python Package for Sampling-based Motion Planner and Samplers
  • Tin Lai
  • Computer Science
    J. Open Source Softw.
  • 2021
TLDR
Sampling-based motion planners are robust methods that avoid explicitly constructing the often intractable high-dimensional configuration space (C-Space).
sbp-env: Sampling-based Motion Planners' Testing Environment
TLDR
Sampling-based motion planners are robust methods that avoid explicitly constructing the often intractable high-dimensional configuration space (C-Space).
A Survey of Asymptotically Optimal Sampling-based Motion Planning Methods
TLDR
This survey summarizes the assumptions behind these popular asymptotically optimal techniques and provides an introduction to the significant ongoing research on this topic, including formal path-quality guarantees for continuously valued search spaces.
Asymptotically Optimal Sampling-Based Motion Planning Methods
TLDR
This research presents a new approach to motion planning that takes into account a robot's limitations and constraints and develops a scalable approach to solve the problem of path planning.
Learning to Plan Optimally with Flow-based Motion Planner
TLDR
This work introduces a conditional normalising flow based distribution that can be conditioned on the current problem instance to provide an informative prior for sampling configurations within promising regions and provides a formulation and theoretical foundation to efficiently sample from the distribution.
...
1
2
...

References

SHOWING 1-10 OF 29 REFERENCES
Statistical Analysis of Circular Data
TLDR
This book presents a meta-modelling framework for analysing two or more samples of unimodal data from von Mises distributions, and some modern Statistical Techniques for Testing and Estimation used in this study.
Balancing Global Exploration and Local-connectivity Exploitation with Rapidly-exploring Random disjointed-Trees
TLDR
This work argues that the active balancing of global exploration and local exploitation is the key to improving sample efficient in sampling-based motion planners, and provides rigorous proofs of completeness and optimal convergence for this novel approach.
A Learning-based Multi-RRT Approach for Robot Path Planning in Narrow Passages
TLDR
Simulation and experimental results show the effectiveness of the proposed LM-RRT approach in single-query path planning problems with narrow passages and its probabilistic completeness and combinatory optimality are proved based on the geometric characteristics of the configuration space.
Learning Sampling Distributions for Robot Motion Planning
TLDR
This paper proposes a methodology for nonuniform sampling, whereby a sampling distribution is learned from demonstrations, and then used to bias sampling, resulting in an order of magnitude improvement in terms of success rate and convergence to the optimal cost.
Sampling-Based Robot Motion Planning: A Review
TLDR
The state of the art in motion planning is surveyed and selected planners that tackle current issues in robotics are addressed, for instance, real-life kinodynamic planning, optimal planning, replanning in dynamic environments, and planning under uncertainty are discussed.
SR-RRT: Selective retraction-based RRT planner
TLDR
A novel retraction-based planner, selective retractions-based RRT, for efficiently handling a wide variety of environments that have different characteristics, and proposes a non-colliding line-test, a dual operator to the bridge line- test, as a culling method to avoid generating samples near wide-open free spaces and thus to generate more samples around narrow passages.
Sampling-based algorithms for optimal motion planning
TLDR
The main contribution of the paper is the introduction of new algorithms, namely, PRM and RRT*, which are provably asymptotically optimal, i.e. such that the cost of the returned solution converges almost surely to the optimum.
Incremental Sampling-based Algorithms for Optimal Motion Planning
TLDR
A new algorithm is considered, called the Rapidly-exploring Random Graph (RRG), and it is shown that the cost of the best path returned by RRG converges to the optimum almost surely, and a tree version of RRG is introduced, called RRT∗, which preserves the asymptotic optimality ofRRG while maintaining a tree structure like RRT.
Triple RRTs: An Effective Method for Path Planning in Narrow Passages
TLDR
A novel two-stage approach to address the problem of multi-d.o.f. robot path planning in high-dimensional configuration space with narrow corridors that keeps a fine balance between global heuristics and local connection, resulting in high performance over the previous RRTs-based path planning methods.
...
1
2
3
...