Abstract:
In this paper, we present a complete and efficient solution of guidance, navigation and control for a quadrotor platform to accomplish 3D coverage flight missions in mapped vineyard terrains. Firstly, an occupancy grid map of the terrain is used to generate a safe guiding coverage path using an Iterative Structured Orientation planning algorithm. Secondly, way-points are extracted from the generated path and added to them trajectory\'s velocities and accelerations constraints. The constrained way-points are fed into a Linear Quadratic Regulator algorithm so as to generate global minimum snap optimal trajectory while satisfying both the pointing and the corridor constraints. Then, when facing unexpected obstacles, the quadrotor tends to re-plan its path in real-time locally using an Improved Artificial Potential Field algorithm. Finally, a geometric trajectory tracking controller is developed on the Special Euclidean group SE(3). The aim of this controller is to track the generated trajectory while pointing towards predetermined direction using the vector measurements provided by the inertial unit. The performance of the proposed method is demonstrated through several simulation results. In particular, safe guiding paths are achieved. Obstacle-free optimal trajectories that satisfy the way-point position, the pointing direction, and the corridor constraints, are successfully generated with optimized platform snap. Besides, the implemented geometric controller can achieve higher trajectory tracking accuracy with an absolute value of the maximum error in the order of 10-3 m.
摘要:
在本文中,我们提出了一个完整而有效的指导解决方案,四旋翼平台的导航和控制,以在绘制的葡萄园地形中完成3D覆盖飞行任务。首先,使用迭代结构化方向规划算法,使用地形的占用网格地图生成安全的引导覆盖路径。其次,从生成的路径中提取路点,并将其添加到轨迹的速度和加速度约束中。受约束的路点被馈送到线性二次调节器算法中,以便在满足指向和走廊约束的同时生成全局最小快照最佳轨迹。然后,当面对意想不到的障碍时,四旋翼倾向于使用改进的人工势场算法在本地实时重新规划其路径。最后,在特殊欧氏群SE(3)上开发了几何轨迹跟踪控制器。该控制器的目的是使用由惯性单元提供的矢量测量来跟踪所生成的轨迹,同时指向预定方向。通过多个仿真结果证明了该方法的性能。特别是,实现了安全的引导路径。满足航路点位置的无障碍最佳轨迹,指向的方向,和走廊的限制,使用优化的平台快照成功生成。此外,