低速车辆在非结构化道路中的轨迹规划
Trajectory Planning of Low-Speed Vehicles on Unstructured Roads

随着汽车保有量的不断增加，导致停车场数量逐渐减少。对于新手驾驶员而言，难以熟练地完成泊车活动。基于此现象，自动泊车技术得到了迅速发展。论文主要对自动泊车的轨迹规划算法进行了研究，通过联立最优控制问题中的车辆运动学约束、起始点约束、车辆速度及加速度约束以及代价函数，完成轨迹规划命题的建模。在命题求解时，选用全联立正交配置有限元法将其离散化为NLP问题，并对代价函数参数进行优化。通过Matlab和AMPL联合仿真求解得出符合约束条件的变量，并使得代价函数最小。实验结果表明，代价函数的优化方法使得车辆在泊车时具有更小的车辆运动前轮转角角速度变化，具备更优的行驶平顺性。
With the increasing number of cars, the number of parking lots is gradually decreasing. It is difficult for novice drivers to skillfully complete parking activities. Based on this phenomenon, automatic parking technology has developed rapidly. This paper mainly studies the trajectory planning algorithm of automatic parking, and completes the modeling of trajectory planning proposition through simultaneous constraints of vehicle kinematics, starting point, vehicle speed and acceleration and cost function in the optimal control problem. When solving the proposition, the fully simultaneous orthogonal collocation finite element method is used to discretize it into NLP problem, and the parameters of the cost function are optimized. Through the joint simulation of Matlab and AMPL, the variables that meet the constraint conditions are obtained and the cost function is minimized. The experimental results show that the optimization method of cost function makes the vehicle have less angular velocity change of front wheel angle and better ride comfort when parking.