HIRO: Heuristics Informed Robot Online Path Planning Using Pre-computed Deterministic Roadmaps

TL;DR

Dividing robot environments into static and dynamic elements, the static part is used for initializing a deterministic roadmap, which provides a lower bound of the final path cost as informed heuristics for fast path-finding as well as guiding a search tree to explore the roadmap during runtime.

Abstract

With the goal of efficiently computing collision-free robot motion trajectories in dynamically changing environments, we present results of a novel method for Heuristics Informed Robot Online Path Planning (HIRO). Dividing robot environments into static and dynamic elements, we use the static part for initializing a deterministic roadmap, which provides a lower bound of the final path cost as informed heuristics for fast path-finding. These heuristics guide a search tree to explore the roadmap during runtime. The search tree examines the edges using a fuzzy collision checking concerning the dynamic environment. Finally, the heuristics tree exploits knowledge fed back from the fuzzy collision checking module and updates the lower bound for the path cost. As we demonstrate in real-world experiments, the closed-loop formed by these three components significantly accelerates the planning procedure. An additional backtracking step ensures the feasibility of the resulting paths. Experiments in simulation and the real world show that HIRO can find collision-free paths considerably faster than baseline methods with and without prior knowledge of the environment.

Figures (6)

• Figure 1: Our algorithm (HIRO, green) compared to LazyPRMbohlin2000path (blue)
• Figure 2: Safe zones (green) of two configurations (black dots) with two degrees of freedom; pink and blue quadrilaterals represent generic constraints determined by different links and obstacles; the green quadrilaterals, i.e., the safe zones, are formed by the strictest constraints from the pink and blue ones. The middle point of the part excluded by the safe zones is the next point on edge to check, marked as red.
• Figure 3: Examples of the planning scene datasets with 4, 8, 12 and 16 spherical random obstacles; start(orange) and goal(blue) poses are randomly generated.
• Figure 4: Improvements regarding the average planning time for different planning scenarios. RRTConnect is used as the baseline. Improvement above one implies that HIRO is better.
• Figure 5: Histogram of the improvements regarding the average planning time for different planning scenarios. Dataset with 4, 8, 12 and 16 are shown in blue, orange, green and red.