Simultaneous dynamic optimization: A trajectory planning method for nonholonomic car-like robots

Two-Stage Trajectory Optimization for Autonomous Ground Vehicles Parking Maneuver

“…Remark 1. It should be noted that in [15], the authors achieved the collision-free with respect to and by transforming these two points to the vehicle's body frame. However, this intuitive restriction of the slot corner might introduce discontinuity, which will have negative effects for the optimization solver.…”

“…This subsection presents a comparative study in terms of the optimal parking movement achieved by performing the proposed two-stage strategy and other optimization-based trajectory planning techniques reported in the literature. For instance, an IPM-based direct transcription method outlined in [15] and an artificial bee colony-based (ABC) intelligent optimization method studied in [29]. For the purpose of comparison, the default setting suggested in the original paper is used.…”

“…The core aim of this problem is to generate a path/trajectory, for the given autonomous vehicle, to reach the specified parking area in the shortest time without colliding other vehicles/obstacles in the environment. Although extensive research work has been carried out on the design of trajectories for different mission profiles and many effective methods were successfully developed for producing the path, it is only in the recent five years that there has been a growing interest in planning trajectories via optimization-based strategies [14], [15]. The motivation for the use of this kind of technique is that in many real-world trajectory design scenarios, not only the path feasibility should be preserved but also various performance indices are desired to be optimized.…”

Influence of the Friction Coefficient on the Trajectory Performance for a Car‐Like Robot

“…This approach marks a clear difference with other planners that only include kinematic constraints, as in Simba et al [15,16]. Those planners either are conservative or do not guarantee the feasibility of the trajectories, as in Li and Shao [17] and Tokekar et al [18].…”

An Inverse Vehicle Model for a Neural-Network-Based Integrated Lateral and Longitudinal Automatic Parking Controller

“…In the geometric planning methods, some form of curves such as β-spline [3], Bézier [4,5], clothoid [6], or polynomial curves [7,8] are generated as a set of reference trajectories. The mathematical optimization approaches, firstly formulating a cost function of the automatic parking problem and then minimizing the objective cost function [9][10][11][12][13]. Besides car-like vehicles, reference trajectory is feasible in the case of parking N-trailer vehicles when complicated kinematic constraints are taken into account in planning the trajectory [14][15][16].…”