Informed sampling space driven robot informative path planning.

Path planning is an important primitive in robotics. In this paper, a new Informed Sampling Space (ISS) driven Informative Path Planning (IPP) approach is developed to facilitate autonomous robots to navigate and explore unknown and hazardous environments for in-situ resource utilization efficiently. The developed ISS-driven IPP approach is targeted on multi-objective optimization enabling the robot to plan its path from start to target locations in the environment and simultaneously explore multiple high-interest areas efficiently. The high-interest areas could be locations advised by a human supervisor or from the robot's prior knowledge of the environment. Typically, a cost function (time, distance, etc.) is used in sampling-based path planners. A new cost function is also developed to incorporate the high-interest spots in this paper, which is based on Multivariate normal (MVN) probability density function (PDF) and a normalization function. Two different IPP models are developed using the new cost function to assist robot navigation. IPP with RRT* is used in the first model with no heuristics, while IPP with RRT* and heuristic ISS is used in the second model. Simulation and comparative analysis substantiate the efficacy and robustness of our approach. The simulation results corroborate that our proposed ISS-driven IPP with RRT* converges rapidly towards the near-optimal solution with respect to both navigation time and environment exploration. • Adaptive cost function incorporates the influence of prior information states. • Multivariate normal probability density function is used to disseminate information. • IPP framework is developed using information map and a sampling-based path planner. • ISS-driven IPP addresses the global sampling problem by ignoring trivial states. • Tree pruning after each subsequent optimization minimizes computational complexity. [ABSTRACT FROM AUTHOR]