Research On Path Planning Of Intelligent Vehicle In And Out Of Garage Based On Direct Collocation Method

With the increase of people’s income and the continuous development of automobile industry,China’s car ownership is growing rapidly.At present,China has entered the era of automobile society,and the production scale and consumption market scale of automobile industry are ranked first in the world.However,the rapid growth of the number of cars has also brought a variety of social problems,such as traffic safety,road congestion,parking space tension and so on.The emergence of intelligent cars to a certain extent solves the above problems.Path planning has always been the focus of intelligent vehicle research,but also the key part of intelligent vehicle technology,with great research significance and application prospects.Intelligent vehicle path planning is to plan a collision free path for the vehicle by using a certain algorithm when the vehicle is in the process of driving and sensing the external environment through sensors.In this paper,according to the functional needs of the intelligent parking lot at this stage,the path planning of the vehicle in and out of the garage is studied.Firstly,p-criterion function is used to describe the boundary of driveable road and static obstacle of various garages.This paper analyzes all kinds of constraints of intelligent vehicle in the local path planning model,including vehicle kinematics constraints,boundary value constraints,path constraints and obstacle avoidance constraints,and establishes the optimal control model of vehicle in and out of the warehouse.By using the direct collocation method,the state variables and control variables in time and space are approximated by fitting polynomials,and the derivative of polynomials is obtained to approximate the state differential quantity at discrete points,so that the optimal control problem is transformed into NLP problem and solved by sequential quadratic programming.In this paper,the access paths of vertical garage,parallel garage and inclined garage are studied respectively,the vehicle kinematics model and garage function are established,various constraints on vehicle operation are analyzed,and the direct collocation method is further used to solve the problem.The results show that the method in this paper can well plan a path for vehicles to avoid obstacles,which has good real-time performance and less delay.This paper compares and analyzes the six working conditions of the intelligent vehiclepath planning based on the direct collocation method.The solution time of each working condition is less than 1.5 seconds,and the solution speed is fast.It shows that the direct collocation method is an effective discrete method for solving the optimal control problem.This method has the advantages of less calculation and better numerical stability in solving the optimal control problem.