Oren Salzman

Asymptotically Near-Optimal RRT for Fast, High-Quality Motion Planning

We present Lower Bound Tree-RRT (LBT-RRT), a single-query sampling-based algorithm that is asymptotically near-optimal. Namely, the solution extracted from LBT-RRT converges to a solution that is within an approximation factor of 1 + ε of the optimal solution. Our algorithm allows for a continuous interpolation between the fast RRT algorithm and the asymptotically optimal RRT* and RRG algorithms. When the approximation factor is 1 (i.e., no approximation is allowed), LBT-RRT behaves like the RRT* algorithm. When the approximation factor is unbounded, LBT-RRT behaves like the RRT algorithm. In between, LBT-RRT is shown to produce paths that have higher quality than RRT would produce and run faster than RRT* would run. This is done by maintaining a tree which is a sub-graph of the RRG roadmap and a second, auxiliary tree, which we call the lower-bound tree. The combination of the two trees, which is faster to maintain than the tree maintained by RRT*, efficiently guarantee asymptotic near-optimality. We suggest to use LBT-RRT for high-quality, anytime motion planning. We demonstrate the performance of the algorithm for scenarios ranging from 3 to 12 degrees of freedom and show that even for small approximation factors, the algorithm produces high-quality solutions (comparable to RRT*) with little runtime overhead when compared to RRT.

Asymptotically-optimal Motion Planning using lower bounds on cost

Many path-finding algorithms on graphs such as A* are sped up by using a heuristic function that gives lower bounds on the cost to reach the goal. Aiming to apply similar techniques to speed up sampling-based motion-planning algorithms, we use effective lower bounds on the cost between configurations to tightly estimate the cost-to-go. We then use these estimates in an anytime asymptotically-optimal algorithm which we call Motion Planning using Lower Bounds (MPLB). MPLB is based on the Fast Marching Trees (FMT*) algorithm [1] recently presented by Janson and Pavone. An advantage of our approach is that in many cases (especially as the number of samples grows) the weight of collision detection in the computation is almost negligible compared to the weight of nearest-neighbor queries. We prove that MPLB performs no more collision-detection calls than an anytime version of FMT*. Additionally, we demonstrate in simulations that for certain scenarios, the algorithmic tools presented here enable efficiently producing low-cost paths while spending only a small fraction of the running time on collision detection.

Motion planning via manifold samples

We present a general and modular algorithmic framework for path planning of robots. Our framework combines geometric methods for exact and complete analysis of low-dimensional configuration spaces, together with sampling-based approaches that are appropriate for higher dimensions. We suggest taking samples that are entire low-dimensional manifolds of the configuration space. These samples capture the connectivity of the configuration space much better than isolated point samples. Geometric algorithms then provide powerful primitive operations for complete analysis of the low-dimensional manifolds. We have implemented our framework for the concrete case of a polygonal robot translating and rotating amidst polygonal obstacles. To this end, we have developed a primitive operation for the analysis of an appropriate set of manifolds using arrangements of curves of rational functions. This modular integration of several carefully engineered components has lead to a significant speedup over the PRM sampling-based algorithm, which represents an approach that is prevalent in practice.

Sparsification of motion-planning roadmaps by edge contraction

We present Roadmap Sparsification by Edge Contraction (RSEC), a simple and effective algorithm for reducing the size of a motion-planning roadmap. The algorithm exhibits minimal effect on the quality of paths that can be extracted from the new roadmap. The primitive operation used by RSEC is edge contraction-the contraction of a roadmap edge to a single vertex and the connection of the new vertex to the neighboring vertices of the contracted edge. For certain scenarios, we compress more than 98% of the edges and vertices at the cost of degradation of average shortest path length by at most 2%.

Finding a needle in an exponential haystack

We present a sampling-based framework for multi-robot motion planning. which combines an implicit representation of roadmaps for multi-robot motion planning with a novel approach for pathfinding in geometrically embedded graphs tailored for our setting. Our pathfinding algorithm, discrete-RRT (dRRT), is an adaptation of the celebrated RRT algorithm for the discrete case of a graph, and it enables a rapid exploration of the high-dimensional configuration space by carefully walking through an implicit representation of the tensor product of roadmaps for the individual robots. We demonstrate our approach experimentally on scenarios that involve as many as 60 degrees of freedom and on scenarios that require tight coordination between robots. On most of these scenarios our algorithm is faster by a factor of at least 10 when compared to existing algorithms that we are aware of.

Efficient high-quality motion planning by fast all-pairs r-nearest-neighbors

Sampling-based motion-planning algorithms typically rely on nearest-neighbor (NN) queries when constructing a roadmap. Recent results suggest that in various settings NN queries may be the computational bottleneck of such algorithms. Moreover, in several asymptotically-optimal algorithms these NN queries are of a specific form: Given a set of points and a radius r report all pairs of points whose distance is at most r. This calls for an application-specific NN data structure tailored to efficiently answering this type of queries. Randomly transformed grids (RTG) were recently proposed by Aiger et al. [1] as a tool to answer such queries in Euclidean spaces and have been shown to outperform common implementations of NN data structures for this type of queries. In this work we employ RTG for sampling-based motion-planning algorithms and describe an efficient implementation of the approach. We show that for motion planning, RTG allow for faster convergence to high-quality solutions when compared to existing NN data structures. Additionally, RTG enable significantly shorter construction times for batched-PRM variants; specifically, we demonstrate a speedup by a factor of two to three for some scenarios.

New perspective on sampling-based motion planning via random geometric graphs

Roadmaps constructed by many sampling-based motion planners coincide, in the absence of obstacles, with standard models of random geometric graphs (RGGs). Those models have been studied for several decades and by now a rich body of literature exists analyzing various properties and types of RGGs. In their seminal work on optimal motion planning Karaman and Frazzoli (2011) conjectured that a sampling-based planner has a certain property if the underlying RGG has this property as well. In this paper we settle this conjecture and leverage it for the development of a general framework for the analysis of sampling-based planners. Our framework, which we call localization-tessellation, allows for easy transfer of arguments on RGGs from the free unit-hypercube to spaces punctured by obstacles, which are geometrically and topologically much more complex. We demonstrate its power by providing alternative and (arguably) simple proofs for probabilistic completeness and asymptotic (near-)optimality of probabilistic roadmaps (PRMs). Furthermore, we introduce several variants of PRMs, analyze them using our framework, and discuss the implications of the analysis.

Motion Planning via Manifold Samples

We present a general and modular algorithmic framework for path planning of robots. Our framework combines geometric methods for exact and complete analysis of low-dimensional configuration spaces, together with practical, considerably simpler sampling-based approaches that are appropriate for higher dimensions. In order to facilitate the transfer of advanced geometric algorithms into practical use, we suggest taking samples that are entire low-dimensional manifolds of the configuration space that capture the connectivity of the configuration space much better than isolated point samples. Geometric algorithms for analysis of low-dimensional manifolds then provide powerful primitive operations. The modular design of the framework enables independent optimization of each modular component. Indeed, we have developed, implemented and optimized a primitive operation for complete and exact combinatorial analysis of a certain set of manifolds, using arrangements of curves of rational functions and concepts of generic programming. This in turn enabled us to implement our framework for the concrete case of a polygonal robot translating and rotating amidst polygonal obstacles. We show that this instance of the framework is probabilistically complete. Moreover, we demonstrate that the integration of several carefully engineered components leads to significant speedup over the popular PRM sampling-based algorithm, which represents the more simplistic approach that is prevalent in practice.

Motion Planning for Multilink Robots by Implicit Configuration-Space Tiling

We study the problem of motion-planning for free-flying multilink robots and develop a sampling-based algorithm that is specifically tailored for the task. Our approach exploits the fact that the set of configurations for which the robot is self-collision free is independent of the obstacles or of the exact placement of the robot. This allows for decoupling between costly self-collision checks on the one hand, which we do off-line (and can even be stored permanently on the robots controller), and collision with obstacles on the other hand, which we compute in the query phase. Our algorithm suggests more flexibility than the prevailing paradigm in which a precomputed roadmap depends both on the robot and on the scenario at hand. We demonstrate the effectiveness of our approach on open and closed-chain multi-link robots, where in some settings our algorithm is more than fifty times faster than commonly used, as well as state-of-the-art solutions.

Optimal motion planning for a tethered robot: Efficient preprocessing for fast shortest paths queries

We study the problem of planning the shortest path for a polygonal robot anchored to a fixed base point by a finite tether translating among polygonal obstacles in the plane. Specifically, we preprocess the workspace to efficiently answer queries of the following type: Given a source location of the robot and an initial configuration of the tether, compute the shortest path to reach a target location while avoiding obstacles and adhering to the tethers constraints. Our work is an extension of the recent work by Kim et al. [1] who considered the problem for a point robot. Their algorithm relies on a discretization of the workspace and is optimal with respect to this discretization. We first replace their grid-based approach with a visibility-graph based approach. This allows to improve the running time of their algorithm by several orders of magnitude. Specifically, testing on a scenario similar to one presented by Kim et al., the running time is improved by a factor of more than 500. Moreover, our approach, which plans optimal paths, is applicable to polygonal (translating) robots and can be used to plan a shortest path while ensuring a predefined clearance from the obstacles. We report on our experimental results on a variety of scenarios. In all cases the preprocessing time is less than one second on a standard-commodity laptop, and a typical query takes several tens of miliseconds.