Motion Primitives-based Path Planning for Fast and Agile Exploration using Aerial Robots

Abstract

This manuscript presents enhancements on our motion-primitives exploration path planning method for agile exploration using aerial robots. The method now further integrates a global planning layer to facilitate reliable largescale exploration. The implemented bifurcation between local and global planning allows for efficient exploration combined with the ability to plan within very large environments, while also ensuring safe and timely return-to-home. A new set of simulation studies and experimental results are presented to demonstrate the new improvements and enhancements. The method is available open-source as a Robot Operating System (ROS) package.

TARE: A Hierarchical Framework for Efficiently Exploring Complex 3D Environments

“…• MBP [24]: A variant of GBP, which constructs the local RRT using motion primitives. The resulting paths are smoother but only span in constrained directions.…”

“…The framework does not involve heuristics, as GBP and MBP, for explicit mode switch. Experiment comparisons to NBVP [21], GBP [23] and MBP [24] show that our method explores much more completely and efficiently while consuming less computation.…”

“…Note that in this method, the exploration mode and relocation mode are explicitly switched using heuristics. Recently, Dharmadhikari et al propose the Motion primitives-Based exploration Planner (MBP) [24], a variant of GBP that constructs the RRT with motion primitives, which produces smoother local paths spanning in constrained directions. In essence, these methods adopt greedy strategies.…”

Hypergame-based Adaptive Behavior Path Planning for Combined Exploration and Visual Search

“…Among others, autonomous robots, be it flying, ground, or underwater, are tasked to explore, map, search and characterize diverse environments of increasing size and complexity. In response to the needs of such applications, the research community has contributed a set of methods for exploration path planning, visual search, and broadly, information sampling [1][2][3][4][5][6][7][8][9][10][11][12][13][14][15][16]. Despite the progress, a limiting factor of the majority of such works relates to the fact that they are optimizing a single objective -for example relating to volumetric exploration [10] or mutual information maximization [14].…”

“…Early work includes the sampling of "next-best-views" [8], and frontiers-based exploration [9]. More recent efforts have proposed receding horizon multi-objective planning [5,6,20,21], graph-based and motion primitives-based methods [10,11], visual search [15,16], information-theoretic schemes [14], and multi-robot strategies [22,23]. Existing planning methods have performed well in simple missions yet cannot provide resilient performance in environments of very large scale, diverse and challenging geometries, and with visually-degraded conditions.…”