月球探测器VTVL避障轨迹快速凸优化

Convex Optimization Based VTVL Trajectory Rapid Optimization for Lunar Rover in the Presence of Terrain Obstacle

针对月球探测器在有月面地形障碍时的垂直起飞-垂直降落(Vertical Takc-off Vertical Landing,VTVL)飞行轨迹优化问题,基于凸优化方法提出一种避障轨迹优化算法。首先考虑没有地形障碍约束时的VTVL轨迹优化问题,利用无损凸化技术将其转化为凸优化问题,快速求解获得初始轨迹,在此基础上,利用凸的椭球面将地形障碍转化为避障约束,并基于初始轨迹对该约束进行线性化处理,形成序列凸优化问题进行求解。数值计算结果表明,本文所提方法能快速高效求解具有地形障碍约束的VTVL避障轨迹优化问题。

A rapid obstacle-avoidance trajectory planning method based on convex optimization is proposed for a lunar rover,which aims at the surface of the moon in a manner of vertical take-off and vertical landing( VTVL) in the presence of terrain obstacle. A VTVL trajectory optimization problem without terrain constraints is firstly considered,and the lossless convexification techniques are used to convert the original problem into an equivalent convex version which can be rapidly solved for having an initial optimal trajectory. Then,the terrain constraint is modeled as a convex ellipsoidal surface and linearized on the initial trajectory therefore by transferring into a sequential convex optimization problem,which is iteratively solved by using convex algorithm. Numerical results reveal that the proposed method is efficiently rapid and effective for solving the VTVL trajectory planning with terrain obstacle constraints.