Dynamic Lane-Changing Trajectory Planning for Autonomous Vehicles Based on Discrete Global Trajectory

Abstract

Automatic lane-changing is a complex and critical task for autonomous vehicle control. Existing researches on autonomous vehicle technology mainly focus on avoiding obstacles; however, few studies have accounted for dynamic lane changing based on some certain assumptions, such as the lane-changing speed is constant or the terminal state is known in advance. In this study, a typical lane-changing scenario is developed with the consideration of preceding and lagging vehicles on the road. Based on the local trajectory generated by the global positioning system, a path planning model and a speed planning model are respectively established through the cubic polynomial interpolation. To guarantee the driving safety, passenger comfort and vehicle efficiency, a comprehensive trajectory optimization function is proposed according to the path planning model and speed planning model. In addition, a dynamic decoupling model is established to solve the problems of real-time application to provide viable solutions. The simulations and real vehicle validations are conducted, and the results highlight that the proposed method can generate a satisfactory lane-changing trajectory for automatic lane-changing actions. Index Terms-Autonomous vehicle, dynamic lane-changing, trajectory planning, real-time optimization, discrete global trajectory NOMENCLATURE A. ACRONYMS ADAS advanced driving assistant systems PFP potential field projection method RRT rapidly exploring random tree TP preceding vehicle on the target lane TL lagging vehicle on the target lane P preceding vehicle on the initial lane E the ego vehicle GPS global positioning system NGSIM next generation simulation SLC the static lane-changing DOLC dynamic optimized lane-changing

QPNet: Lane‐changing trajectory planning combining quadratic programming and neural network under the convex optimization framework

“…The proposed QPNet model addresses the shortcomings of the existing lane-changing trajectory planning model such as long model CPU solution time, strong assumptions on the traffic states of surrounding vehicles, only planning lane-changing trajectory on straight and etc [33,36,55]. the model CPU solution time plays a significant role in reacting to the changes in the driving environments.…”

“…This method plans the optimal lane-changing trajectory while satisfying the safety and comfort constraints; the solution is suitable for planning in the dynamic environment; but usually involves nonconvex optimization problems, so that the calculation cost is high and cannot meet real-time requirement. [17,[36][37][38] Lane It should be noted that our model belongs to the latter. For the former study, the purpose is to maximize the restoration of the real lane-changing behaviour by building a model.…”

“…This problem led to the development of the mathemat-ical programming planner, which converts the lane-changing trajectory planning problem into a constrained optimization mathematical problem. Liu et al [36] presented a dynamic decoupling model based on mathematical optimization. They generated the lane-changing path and speed profiles through the cubic polynomial interpolation, and then designed a comprehensive trajectory optimization function to ensure the safety and comfort of the vehicles.…”

“…Moreover, 𝜆Δa 2 0,I is used to guarantee that the final acceleration of the ego vehicle a 0,I tends to the car-following acceleration, so that the ego vehicle can switch to the car-following manoeuvre comfortably after lane-changing. 𝜆 is set as 1000, and Δa 0,I is calculated using Equation (36), where the values of parameters and the definitions of variables have already described in Section 4.3.3. By solving for the minimum value of the objective function, the optimal lane-changing velocity profiles that meet the constraints can, therefore, be planned.…”

“…Liu et al. [36] presented a dynamic decoupling model based on mathematical optimization. They generated the lane‐changing path and speed profiles through the cubic polynomial interpolation, and then designed a comprehensive trajectory optimization function to ensure the safety and comfort of the vehicles.…”