To control autonomous vehicles (AVs) in urban unsignalized intersections is a challenging problem, especially in a hybrid traffic environment where self-driving vehicles coexist with human driving vehicles. In this study, a coordinated control method with proximal policy optimization (PPO) in Vehicle-Road-Cloud Integration System (VRCIS) is proposed, where this control problem is formulated as a reinforcement learning (RL) problem. In this system, vehicles and everything (V2X) was used to keep communication between vehicles, and vehicle wireless technology can detect vehicles that use vehicles and infrastructure (V2I) wireless communication, thereby achieving a cost-efficient method. Then, the connected and autonomous vehicle (CAV) defined in the VRCIS learned a policy to adapt to human driving vehicles (HDVs) across the intersection safely by reinforcement learning (RL). We have developed a valid, scalable RL framework, which can communicate topologies that may be dynamic traffic. Then, state, action and reward of RL are designed according to urban unsignalized intersection problem. Finally, how to deploy within the RL framework was described, and several experiments with this framework were undertaken to verify the effectiveness of the proposed method.

In this study, we present and discuss a variant of the classical vehicle routing problem (VRP), namely the heterogeneous multitype fleet green automated guided vehicle (AGV) routing problem with time windows (HFGVRPTW) applied in the workshops of flexible manufacturing systems (FMS). Specifically, based on the analysis of AGV body structure and motion state, transport distance and energy consumption are selected as two optimization objectives. According to the characteristics and application context of the problem, this paper designs a hybrid genetic algorithm with large neighborhood search (GA-LNS) considering the farthest insertion heuristic. GA-LNS is improved by increasing the local search ability of genetic algorithm to enhance the solution optimal quality. Extensive computational experiments which are generated from Solomon’s benchmark instances and a real case of FMS are designed to evaluate and demonstrate the efficiency and effectiveness of the proposed model and algorithm. The experimental results reveal that compared with using the traditional homogeneous fleet, the heterogeneous multitype AGV fleet transportation mode has a huge energy-saving potential in workshop intralogistics.

This paper addresses a collaborative multi-carrier vehicle routing problem (CMCVRP) where carriers tackle their orders collaboratively to reduce transportation costs. First, a hierarchical heuristics algorithm is proposed to solve the transportation planning problem. This algorithm makes order assignments based on two distance rules and solves the vehicle routing problem with a hybrid genetic algorithm. Second, the profit arising from the coalition is quantified, and an improved Shapley value method is proposed to distribute the profit fairly to individual players. Extensive experiment results showed the effectiveness of the proposed hierarchical heuristics algorithm and confirmed the stability and fairness of the improved Shapley value method.

The signal control system for multi-intersection traffic management is an extensive complex nonlinear and non-equilibrium system and usually leads to a weak control strategy. Existing studies of traffic dynamics mainly focus on the applications to the single network, whereas how the structure of multiple intersections affects the dynamical behaviors is unclear. Connected vehicles in recent studies have shown the potential to eliminate traffic communication barriers and balance the traffic load. For this consideration, we propose a new multi-intersection model based on Vehicle-to-Everything (V2X) network to capture the features of traffic flow, which employs vehicular communication to obtain real-time traffic information and share it among the adjacent intersections. Focus on smoothing traffic flow without stop-and-go driving and improving the coordination of multiple traffic lights, we develop a multi-intersection coordination algorithm based on V2X (MICA-V for short) to maximize the green waveband in the arterial road. Due to the establishment of registration list yielding the average saturated headway time, our method is of high efficiency—the distributed process of data makes the convergence to be very fast with nearly linear time. Simulation results via MATLAB with Simulation of Urban Mobility (SUMO) showed the effectiveness of the proposed MICA-V in improving traffic capacity, reducing the number of stop behaviors, and improve the average speed of vehicles moving on the arterial road.

Recent advances in distributed machine learning and wireless network technologies are bringing new opportunities for Internet of Things (IoT) systems, where smart devices are often wirelessly connected to collaborate, jointly completing tasks known as communication-dependent computing (CDC) tasks. However, due to the dependence of computing on communication and the presence of concurrent tasks, it remains a challenge to optimize CDC task performance and efficiency while fulfilling multi-dimensional requirements, particularly with incomplete system information and dynamic environmental impacts. To overcome these, we present a concurrent CDC task framework to model the correlated communication and computing stages and multi-dimensional requirements of CDC tasks. We then formulate a task orchestration and resource management problem to optimize overall utility, where each task's utility is designed as a joint metric including the cumulative computing deviation and time efficiency of task completion. To solve this, we employ auxiliary graphs to capture the topological information of tasks and resources, and update weights based on the utility in dynamic environments. Subsequently, a multi-agent reinforcement learning algorithm is leveraged to make distributed decisions with incomplete information. Experiments demonstrate the proposed approach outperforms baselines in terms of task performance and efficiency, indicating our solution holds great potential.