In this paper, we study the difference between two major strategies of cooperative driving at non-signalized intersections: namely the “ad hoc negotiation based” strategy and the "planning based" strategy. The fundamental divide of these two strategies lies in how to determine the passing order of vehicles at intersections. The "ad hoc negotiation based" strategy makes vehicles roughly follow first-come-first-served order but allows some adjustments. This leads to a local optimal solution in many situations. The "planning based" strategy aims to find a global optimal passing order of vehicles. However, constrained by the planning complexity and time requirement, we often stop at a local optimal solution, too. We carry out a series of simulations to compare the solutions found by two strategies, under different traffic scenarios. Results indicate the performance of a strategy mainly depends on the passing order of vehicles that it finds. Although there exist several trajectory planning algorithms associating with the solving process of passing orders, their differences are negligible. Moreover, if the traffic demand is very low, the performance difference between two strategies is small. When the traffic demand becomes high, the "planning based" strategy yields significantly better performance since it finds better passing orders. These findings are important to cooperative driving study.

In this paper, we study how to learn a harmonious deep reinforcement learning (DRL) based lane-changing strategy for autonomous vehicles without Vehicle-to-Everything (V2X) communication support. The basic framework of this paper can be viewed as a multi-agent reinforcement learning in which different agents will exchange their strategies after each round of learning to reach a zero-sum game state. Unlike cooperation driving, harmonious driving only relies on individual vehicles’ limited sensing results to balance overall and individual efficiency. Specifically, we propose a well-designed reward that combines individual efficiency with overall efficiency for harmony, instead of only emphasizing individual interests like competitive strategy. Testing results show that competitive strategy often leads to selfish lane change behaviors, anarchy of crowd, and thus the degeneration of traffic efficiency. In contrast, the proposed harmonious strategy can promote traffic efficiency in both free flow and traffic jam than the competitive strategy. This interesting finding indicates that we should take care of the reward setting for reinforcement learning-based AI robots (e.g., automated vehicles) design, when the utilities of these robots are not strictly in alignment.

One of the most challenging problems in human-machine co-work is the gap between human intention and the machine's understanding and execution. Large Language Models (LLMs) have been showing superior abilities in solving such issue. In this paper, we design a universal framework that embeds LLMs as a vehicle “Co-Pilot” of driving, which can accomplish specific driving tasks with human intention satisfied based on the information provided. Meanwhile, a utilization workflow is defined to handle the interaction between humans and vehicles, and memory mechanism is introduced to organize the information involved in the tasks. Expert Oriented Black-Box tuning is proposed to improve the performance of the Co-Pilot without finetuning or training the LLMs. In the experiment, the Co-Pilot is applied to two different tasks, i.e., path tracking control and trajectory planning. The Co-Pilot adjusts vehicle operating conditions by selecting a proper controller or planning a certain trajectory to fit human intentions. Simulation tests are conducted to evaluate the performance and generality of the proposed module. The results show that the Co-Pilot can accomplish most of the tasks based on only natural language processing, although it is not flawless. Finally, a discussion about human-machine hybrid intelligence and further applications of LLMs in autonomous driving is made. We believe that such a framework has promising potential in further applications in the field of automous vehicles.

There are many constraints to be considered to plan a safe and feasible trajectory for autonomous mining trucks due to the rough terrain in open-pit mines. It is difficult to handle these constraints uniformly within a trajectory planning framework. To overcome this limitation, a novel trajectory planning framework is proposed in this paper. Different from many existing studies which characterized the spatial positions of vehicles at the sampling time points as decision variables, we characterize the time points at the sampling spatial positions as decision variables to conveniently address the terrain constraints. Based on such a formulation, we further consider the kinematic constraints, engine power constraints and time window constraints. Various objectives (e.g., transport efficiency, fuel economy) can be considered in this trajectory planning framework, and the formulated trajectory planning problems can be efficiently solved. Finally, several case studies are conducted to validate the effectiveness of the proposed trajectory planning framework.

Recent studies have made dozens of attempts to apply multi-agent deep reinforcement learning (MARL) for large-scale traffic signal control. However, most related studies have ignored how to master arterial traffic signal control. We cannot easily extract useful information and search solution space because the arterial traffic control problem has large state-action spaces. Here we tackle these issues by proposing a multi-agent attention-base soft actor-critic (MASAC) model to master arterial traffic control. Specifically, we implement the attention mechanism in the actor and critic network to enhance traffic information extraction ability. More importantly, we are the first to apply the soft actor-critic (SAC) algorithm to train the arterial traffic control model to search more solution spaces. Testing results indicate that the MASAC method significantly outperforms existing MARL algorithms and the multiband-based method. These findings can help researchers to design better model structures for other MARL problems.