LP-based Optimal Path Planning in Acceleration Space
Di Zu, Jianda Han, Dalong Tan · 2006
This paper proposed Acceleration Space Linear Programming (ASLP) method for the path planning of target pursuit and obstacles avoidance (TPOA) problem in dynamic and uncertain environment. The acceleration space for TPOA problem is defined and LP algorithm is introduced into it. LP is a powerful tool for optimization but exhibits weakness when there exist nonlinear constraints or objective function. In the proposed acceleration space, the constraints and objective function of TOPA problems are inherently described as linear inequalities that can be directly solved by LP. Also, the complexity in obstacle contour does not increase the computational burden of the ASLP. Simulations are conduced with respect to the TPOA scenario including one moving target and multiple moving obstacles. The results obtained by ASLP are presented and compared with those of artificial potential guided evolution algorithm (APEA). Significant superior is demonstrated in many aspects by the comparison.