Intelligent Transportation Vehicle Road Collaboration and Task Scheduling Based on Deep Learning in Augmented Internet of Things

With the continuous development of intelligent transportation systems, achieving efficient vehicle-road coordination and task scheduling has become increasingly important for enhancing safety and efficiency. In order to effectively solve the problem that it is difficult to adapt to the intelligent transportation system due to the heterogeneity of cloud and edge computing resources and the complexity of communication, this paper proposes an intelligent transportation vehicle road collaboration and task scheduling method based on deep learning in the enhanced Internet of Things (IoT) environment. The proposed method utilizes a Collaborative Task Placement Deep Reinforcement Learning strategy (CTPDRL) and builds a cloud-edge collaborative computing framework controlled by deep reinforcement learning. Then, from the perspective of the interests of users and service providers, the task model under cloud edge collaborative computing is analyzed, and a system service quality model is established. Through Deep Q Networks (DQN) and Q-tables, CTPDRL can optimize the allocation of computing and communication resources, achieving effective task placement. The experimental results show that our proposed method can effectively reduce the overall cost of cloud and edge cloud computing, improve the practicality of the system in multi edge cloud environments, and help solve the problem of image perception blind spots in intelligent transportation, thereby improving the safety and efficiency of intelligent driving.

I. Introduction

Vehicle road collaboration technology is built on advanced communication technology, and the implementation of various applications is based on information exchange between vehicles and the surrounding environment. The reliability, effectiveness, and stability of information transmission are crucial for smart transportation. Due to the fact that vehicle road collaboration is a dynamic and open system, its participants include vehicles, roadside devices, pedestrians, etc [1]. Task scheduling is a core function of vehicle road collaboration, including traffic signal control, vehicle navigation, emergency response, and other aspects. Through deep learning models, the system can adjust task priorities in real-time and dynamically allocate resources. For example, by analyzing real-time traffic data, deep learning models can predict possible congestion at a certain intersection, adjust the timing of traffic lights, and optimize traffic flow. The deep learning model in the vehicle road collaborative system can continuously optimize its own parameters based on real-time data, thus possessing stronger adaptability to environmental changes. The information exchange in vehicle road collaboration is different from general wireless communication. The environment it operates in is complex and ever-changing, and the occlusion of vehicles and buildings in urban road environments can also bring great uncertainty to the interaction, which poses challenges to the reliability and stability of vehicle road collaborative information transmission [2]. Therefore, information transmission in vehicle road collaboration is a very complex problem, and building a reliable and safe information exchange environment for vehicle road collaboration is crucial for intelligent transportation.