Vision-based curved lane keeping control for intelligent vehicle highway system

Abstract

This article addresses the coordinated longitudinal and lateral motion control for an intelligent vehicle highway system. The strategy of this work consists of defining the edges of the traveled lane using a vision sensor. According to the detected boundaries, a constrained path-following method is proposed to drive the longitudinal and the lateral vehicle’s motion. Error constraints of the intelligent vehicle highway system position are manipulated by including the function of barrier Lyapunov in designing the guidance algorithm for the intelligent vehicle highway system. To calculate the necessary forces that would steer the vehicle to the desired path, a control design is proposed that integrates the sign of the error for the compensation of the uncertain vehicle’s parameters. The Lyapunov function is later used to minimize the path-following errors and to guarantee a stable system. The efficiency of the developed approach is proved by numerical simulations.

Longitudinal and lateral control of autonomous vehicles in multi‐vehicle driving environments

“…Note that any trajectory should be confined to the vehicle dynamics limits [28, 30]. Thus, the stable handling envelope is first derived from the vehicle 2‐DOF dynamic model to extract the stable trajectories from the whole trajectory cluster.…”

“…In [27], a non‐linear feedback control was proposed for the non‐linear vehicle dynamics with varying velocity, where the stabilisation of longitudinal, lateral and yaw angular vehicle velocities was guaranteed simultaneously. The article [28] investigated coordinated longitudinal and lateral motion control for AVs. Specifically, a barrier Lyapunov‐based sliding mode control (SMC) was developed to satisfy error constraints and provide uncertainty compensation.…”

Design and Implementation of HD Mapping, Vehicle Control, and V2I Communication for Robo-Taxi Services

“…All the factors used to calculate the points are listed with their corresponding weights (see Table 1). Accordingly, the vehicles are ranked according to the points that they accumulated during the contest, which lasted for 1.5 h. The following core competencies were required for vehicles in the contest to demonstrate the robo-taxi service: mapping; localization [10][11][12][13][14][15][16]; path-planning [17,18]; vehicle control [19][20][21][22][23][24]; and perception [25][26][27][28].…”

Development of a dual-stage and visual-servo filming robot

“…It is a very appropriate method for tracking target in filming industry where the actors are in general not able to mount extra components for other types of sensors to track. Visual servoing is also a popular topic among academia, and the applications of this field are vast, including pick and place, 5,6 state estimation, 7 unmanned aerial vehicles, 8–10 lane keeping, 11 vision-based robots, 12–15 automatic microassembly, 16 and car transportation. 17 In addition, various types of control strategies have been used for visual servoing depending on their applications.…”