Bayesian Local Sampling-Based Planning

Lai, Tin; Morere, Philippe; Ramos, Fábio; Francis, Gilad

doi:10.1109/lra.2020.2969145

Cited by 34 publications

(15 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The bridge test [6] tries to find a collision-free middle point between configurations that are in collision with the obstacles. In [8], a Bayesian learning scheme is used to model sampling distributions, which subsequently updates based on previous samples and maximizes the likelihood of sampling from high probability regions.…”

Section: A the Challenge Of Narrow Passagesmentioning

confidence: 99%

See 1 more Smart Citation

Efficient Path Planning in Narrow Passages for Robots with Ellipsoidal Components

Ruan¹,

Poblete²,

Wu³

et al. 2021

Preprint

View full text Add to dashboard Cite

Path planning has long been one of the major research areas in robotics, with PRM and RRT being two of the most effective classes of path planners. Though generally very efficient, these sampling-based planners can become computationally expensive in the important case of "narrow passages". This paper develops a path planning paradigm specifically formulated for narrow passage problems. The core is based on planning for rigid-body robots encapsulated by unions of ellipsoids. The environmental features are enclosed geometrically using convex differentiable surfaces (e.g., superquadrics). The main benefit of doing this is that configuration-space obstacles can be parameterized explicitly in closed form, thereby allowing prior knowledge to be used to avoid sampling infeasible configurations. Then, by characterizing a tight volume bound for multiple ellipsoids, robot transitions involving rotations are guaranteed to be collision-free without traditional collision detection. Furthermore, combining the stochastic sampling strategy, the proposed planning framework can be extended to solving higher dimensional problems in which the robot has a moving base and articulated appendages. Benchmark results show that, remarkably, the proposed framework outperforms the popular sampling-based planners in terms of computational time and success rate in finding a path through narrow corridors and in higher dimensional configuration spaces.

show abstract

Section: A the Challenge Of Narrow Passagesmentioning

confidence: 99%

“…valid configurations in a narrow passage, various methods have been proposed such as [6]- [8] (Sec. II-A provides more detailed reviews on narrow passage problems).…”

Section: Introductionmentioning

confidence: 99%

Efficient Path Planning in Narrow Passages for Robots with Ellipsoidal Components

Ruan¹,

Poblete²,

Wu³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…There is an extensive literature on estimating complex distributions, but they are not always applicable to motion planning due to issues like runtime performance [21], loss of completeness guarantee [22], or falling into the same common pitfall of mode collapse in generative models [20]. Modelling sampling distribution can improve the efficiency of SBPs without compromising completeness [23], yet it remains an open question how to find a suitable learning-based procedure to effectively utilise prior experience. Encoder networks are a class of promising models that can significantly enhance the sampling efficiency in SBPs [16], yet standard techniques like variational autoencoders [24] are prone to mode collapse due to the target distribution C-space's sparsity.…”

Section: The Quality Of Sampling Distribution Is Critical In Sbps As ...mentioning

confidence: 99%

Learning to Plan Optimally with Flow-based Motion Planner

Lai¹,

Ramos²

2020

Preprint

Self Cite

View full text Add to dashboard Cite

Sampling-based motion planning is the predominant paradigm in many real-world robotic applications, but its performance is immensely dependent on the quality of the samples. The majority of traditional planners are inefficient as they use uninformative sampling distributions as opposed to exploiting structures and patterns in the problem to guide better sampling strategies. Moreover, most current learning-based planners are susceptible to posterior collapse or mode collapse due to the sparsity and highly varying nature of C-Space and motion plan configurations. In this work, we introduce a conditional normalising flow based distribution learned through previous experiences to improve sampling of these methods. Our distribution can be conditioned on the current problem instance to provide an informative prior for sampling configurations within promising regions. When we train our sampler with an expert planner, the resulting distribution is often near-optimal, and the planner can find a solution faster, with less invalid samples, and less initial cost. The normalising flow based distribution uses simple invertible transformations that are very computationally efficient, and our optimisation formulation explicitly avoids mode collapse in contrast to other existing learning-based planners. Finally, we provide a formulation and theoretical foundation to efficiently sample from the distribution; and demonstrate experimentally that, by using our normalising flow based distribution, a solution can be found faster, with less samples and better overall runtime performance.

show abstract

“…R R F * overcomes this limitation by addressing the sampling-based motion planning problem with a divide-and-conquer approach. R R F * follows the approach in [27] which uses local trees for Bayesian local sampling. However, instead of purely using Markov Chain to plan within C-space [27], R R F * adaptively locates difficult regions for tree extensions and only proposes new local trees when local planning is beneficial.…”

Section: Introductionmentioning

confidence: 99%

“…R R F * follows the approach in [27] which uses local trees for Bayesian local sampling. However, instead of purely using Markov Chain to plan within C-space [27], R R F * adaptively locates difficult regions for tree extensions and only proposes new local trees when local planning is beneficial. R R F * uses two rooted trees and reformulate the spawning of local-trees as an adaptive selection process.…”

Section: Introductionmentioning

confidence: 99%

Rapidly-exploring Random Forest: Adaptively Exploits Local Structure with Generalised Multi-Trees Motion Planning

Lai

2021

Preprint

Self Cite

View full text Add to dashboard Cite

Sampling-based motion planners perform exceptionally well in robotic applications that operate in highdimensional space. However, most works often constrain the planning workspace rooted at some fixed locations, do not adaptively reason on strategy in narrow passages, and ignore valuable local structure information. In this paper, we propose 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. Local sampling exploits the local-connectivity of spaces via Markov Chain random sampling, which is updated sequentially with a Bayesian proposal distribution to learns the local structure from past observations. The trees selection problem is formulated as a multi-armed bandit problem, which efficiently allocates resources on the most promising tree to accelerate planning runtime. R R F * learns the region that is difficult to perform tree extensions and adaptively deploys local sampling in those regions to maximise the benefit of exploiting local structure. We provide rigorous proofs of completeness and optimal convergence guarantees, and we experimentally demonstrate that the effectiveness of R R F * 's adaptive multi-trees approach allows it to performs well in a wide range of problems.

show abstract

Bayesian Local Sampling-Based Planning

Cited by 34 publications

References 20 publications

Efficient Path Planning in Narrow Passages for Robots with Ellipsoidal Components

Efficient Path Planning in Narrow Passages for Robots with Ellipsoidal Components

Learning to Plan Optimally with Flow-based Motion Planner

Rapidly-exploring Random Forest: Adaptively Exploits Local Structure with Generalised Multi-Trees Motion Planning

Contact Info

Product

Resources

About