Many task execution techniques tend to repeatedly invoke motion planning algorithms in order to perform complex tasks. In order to accelerate the perform of such methods, we present a real-time global motion planner that utilizes the computational capabilities of current many-core GPUs (graphics processing units). Our approach is based on randomized sample-based planners and we describe highly parallel algorithms to generate samples, perform collision queries, nearest-neighbor computations, local planning and graph search to compute collision-free paths for rigid robots. Our approach can efficiently solve the single-query and multiquery versions of the planning problem and can obtain one to two orders of speedup over prior CPU-based global planning algorithms. The resulting GPU-based planning algorithm can also be used for real-time feedback for task execution in challenging scenarios.
We present a novel optimization-based algorithm for motion planning in dynamic environments. Our approach uses a stochastic trajectory optimization framework to avoid collisions and satisfy smoothness and dynamics constraints. Our algorithm does not require a priori knowledge about global motion or trajectories of dynamic obstacles. Rather, we compute a conservative local bound on the position or trajectory of each obstacle over a short time and use the bound to compute a collision-free trajectory for the robot in an incremental manner. Moreover, we interleave planning and execution of the robot in an adaptive manner to balance between the planning horizon and responsiveness to obstacle. We highlight the performance of our planner in a simulated dynamic environment with the 7-DOF PR2 robot arm and dynamic obstacles.
Motivated by what is required for real-time path planning, the paper starts out by presenting RMPD, a new recursive ''local'' planner founded on the key notion that, unless made necessary by an obstacle, there must be no deviation from the shortest path between any two points, which would normally be a straight line path in the configuration space. Subsequently, we increase the power of RMPD by introducing the notion of cost-awareness into the algorithm to improve the path quality -- this is done by associating obstacle and smoothness costs with the currently selected path points and factoring those costs in choosing the best points for the next iteration. In this manner, the overall strategy in the cost-aware form of RMPD, cRMPD, combines the computational efficiency made possible by the recursive RMPD planner with the cost efficacy of a stochastic trajectory optimizer to rapidly produce high-quality local collision-free paths. Based on the test cases we have run, our experiments show that cRMPD can reduce planning time by up to two orders of magnitude as compared to RRT-Connect, while still maintaining a path length optimality equivalent to that of RRT*.
Sampling-based search has been shown effective in motion planning, a hard continuous state-space problem. Motion planning is especially challenging when the robotic system obeys differential constraints, such as an acceleration controlled car that cannot move sideways. Methods that expand trajectory trees in the state space produce feasible solutions for such systems. These planners can be viewed as continuous-space analogs of traditional uninformed search as their goal is to explore the entire state space. In many cases, the search can be focused on the part of the state-space necessary to solve a problem by employing heuristics.
In (Otte and Correll 2013) we present C-FOREST, a parallelization framework for single-query sampling-based shortest-path planning algorithms. C-FOREST has been observed to have super linear speedup on many problems, e.g., paths of quality Ltarget are found 350X faster by 64 CPUs working in parallel than by 1 CPU. In (Otte and Correll 2013) C-FOREST is tested in conjunction with the RRT* algorithm. In the current work we perform additional experiments that show C-FOREST provides similar advantages when used conjunction with the SPRT algorithm. This reinforces our original claim that C-FOREST is generally applicable to a wide range of sampling based motion planning algorithms.