Trajectory Planning for Autonomous High-Speed Overtaking using MPC with Terminal Set Constraints

Abstract

With self-driving vehicles being pushed towards the mainstream , there is an increasing motivation towards development of systems that autonomously perform manoeuvres involving combined lateral-longitudinal motion (e.g., lanechange, merge, overtake, etc.). This paper presents a situational awareness and trajectory planning framework for performing autonomous overtaking manoeuvres. A combination of a potential field-like function and reachability sets of a vehicle are used to identify safe zones on a road that the vehicle can navigate towards. These safe zones are provided to a model predictive controller as reference to generate feasible trajectories for a vehicle. The strengths of the proposed framework are: (i) it is free from non-convex collision avoidance constraints, (ii) it ensures feasibility of trajectory, and (iii) it is real-time implementable. A proof of concept simulation is shown to demonstrate the ability to plan trajectories for high-speed overtaking manoeuvres.

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

“…Recently, the rapid development of vehicle‐to‐vehicle (V2V) and vehicle‐to‐infrastructure (V2I) communications [18–20] has brought about the possibility of automatic lane change manoeuvre sharing based on information from human drivers, vehicles and the environment. Dixit et al [21] proposed a combined potential field and feasible set framework for situational awareness and trajectory planning for autonomous overtaking. A tube‐based robust MPC was designed to generate a feasible trajectory for longitudinal and lateral motion of a vehicle.…”

“…In this work, under the hierarchical design framework, a novel longitudinal and lateral control of AVs is proposed to satisfy a wide range of lane change manoeuvres in multi‐vehicle driving environments. The main contributions are as follows: In contrast to [21, 25] where autonomous lane change is performed in the simple scenario with a preceding vehicle and two lanes, the proposed trajectory planning algorithm investigates the cases in multi‐vehicle driving environments with more surrounding vehicles and three lanes. Unlike purely using lateral trajectory planning [26] or numerical solution of non‐convex optimisation [22–24], the approach presented here offers the optimal longitudinal and lateral trajectories as polynomials in terms of a uniform parameter that is determined by considering vehicle dynamics stability, collision avoidance and driver preference simultaneously. Different from the existing triple‐step control [33, 38], a MIMO triple‐step non‐linear control strategy is developed to achieve longitudinal and lateral tracking control and guarantee the stability of the closed‐loop system under zero dynamics. The remainder of this paper is organised as follows. Longitudinal and lateral motion planning is described to generate a trajectory cluster in Section 2.…”

Autonomous Collision Avoidance Using MPC with LQR-Based Weight Transformation

“…MPC performance is highly reliant on the weights on each objective but the challenge is that there is no systematic procedure to select the weights to assure the best performance of MPC. Currently, suitable weight selection is a time consuming procedure that requires a large number of trial and error selections and many computer simulations [ 1 , 2 , 3 ]. The complexity increases with an increase in the number of control objectives [ 4 ].…”

Trajectory Tracking Control Algorithm for Autonomous Vehicle Considering Cornering Characteristics

“…The MPC optimization problem was transformed into a parameter nonlinear programming problem through a scenario tree, and the calculation efficiency was raised. Dixit et al [26] used a model predictive controller that can provide a reference trajectory for automatic overtaking of unmanned vehicles, dealt with non-convex collision constraints, and achieved trajectory tracking during high-speed overtaking. In summary, many research methods have gained certain research results around trajectory tracking control, but each method has its advantages and disadvantages, and there are different degrees of limitations.…”