A path planning and path‐following control framework for a general 2‐trailer with a car‐like tractor

Abstract

Maneuvering a general 2‐trailer with a car‐like tractor in backward motion is a task that requires a significant skill to master and is unarguably one of the most complicated tasks a truck driver has to perform. This paper presents a path planning and path‐following control solution that can be used to automatically plan and execute difficult parking and obstacle avoidance maneuvers by combining backward and forward motion. A lattice‐based path planning framework is developed in order to generate kinematically feasible and collision‐free paths and a path‐following controller is designed to stabilize the lateral and angular path‐following error states during path execution. To estimate the vehicle state needed for control, a nonlinear observer is developed, which only utilizes information from sensors that are mounted on the car‐like tractor, making the system independent of additional trailer sensors. The proposed path‐planning and path‐following control framework is implemented on a full‐scale test vehicle and results from simulations and real‐world experiments are presented.

Improved Path Planning by Tightly Combining Lattice-Based Path Planning and Optimal Control

“…After the discretization has been selected, the connectivity in the graph is chosen by selecting which neighboring states to connect. Finally, the set of motion primitives P is constructed by computing N m motion segments (for example using an OCP solver as in [20], [5]) needed to connect the neighboring states, without considering obstacles. A motion primitive m ∈ P is defined as…”

“…In this paper, optimal path planning is defined as the problem of finding a feasible and collision-free path from the vehicle initial state to a desired goal state, while a specified performance measure is minimized. This path is then intended to be used as a reference for a path-following controller, such as the ones described in [2], [4], [5].…”

On Motion Planning Using Numerical Optimal Control

“…To obtain a more reliable convergence for the nlp solver, we have observed that it can be beneficial to compute the motion primitives for backward motion from the goal state to the initial state in forward motion for which the system is stable. Since the truck and trailer system is on the form of Lemma 1 in [52], it is possible, according to Theorem 1 in [52], to obtain an optimal backward motion (x * t,b (τ ), u * b (τ )), τ ∈ [0, T ] that brings the system from x start to x final , from the optimal forward motion (x * t,f (t), u * f (t)), t ∈ [0, T ]. This can be achieved by the state and control transformation: 15) where the optimal forward motion is obtained by solving the motion primitive ocp (4.2) from x final to x start .…”

“…Note that it is only the steering angle rate ω and the velocity v 1 that change signs, while the other states and controls are fully time-reversed. For a detailed description of this property, the reader is referred to [52].…”

Deep Reinforcement Learning Controller for 3D Path Following and Collision Avoidance by Autonomous Underwater Vehicles

“…For road vehicles, several path/motion planning strategies are demonstrated ( Karaman and Frazzoli, 2011 ). In their recent studies, Ljungqvist et al (2019) and Cirillo (2017) demonstrated the use of the Lattice-based method for path/motion planning. In both works where the focus was path planning, the algorithms developed were tested on real vehicles.…”