Adaptive Region Boosting method with biased entropy for path planning in changing environment

摘要: Abstract Path planning in changing environments with difficult regions, such as narrow passages and obstacle boundaries, creates significant challenges. As the obstacles W-space move frequently, crowd degree of C-space changes accordingly. Therefore, order to dynamically improve sampling quality, it is appreciated for a planner rapidly approximate different parts C-space, boost sample densities them based on their difficulty levels. In this paper, novel approach called Adaptive Region Boosting (ARB) proposed increase density areas strategies. What's more, new criterion, biased entropy, evaluate region. The criterion takes into account both temporal spatial information specific region, make thorough assessment local area. Three groups experiments are conducted dual-manipulator system 12 DoFs. Experimental results indicate that ARB effectively improves success rate outperforms all other related methods various dynamical scenarios.