With the widespread application of the new generation of mobile communication technology in automobiles, Intelligent and Connected Vehicles (ICVs) equipped with communication devices are experiencing rapid development. The dynamic and static information of roads and traffic hold significant latent value for various applications, particularly in the domains of vehicle safety, energy-efficient driving, and the efficient operation of traffic. Predictive Cruise Control (PCC) is a technology that optimally plans and controls vehicle movements by fully leveraging future road traffic information. Constrained by issues such as the limited perception range and weak onboard computing capabilities of traditional single-vehicle intelligent driving, the technological iteration and industrial development of PCC have encountered considerable obstacles. In recent years, with the continuous improvement of infrastructure such as the Vehicle-Road-Cloud Integrated Control System, especially the enhancement of roadside perception and big data cloud computing, vehicles have gradually gained the capability for beyond-line-of-sight perception of road traffic information and fused decision-making computation. This has ushered in new opportunities to overcome the technological bottlenecks of single-vehicle intelligence in perception, decision-making, and control, and has swiftly become a significant research direction in current cloud-controlled applications for ICVs. Presently, research on PCC technology predominantly focuses on longitudinal speed control applications of vehicles under single-vehicle intelligence system, with few researchers providing a comprehensive summary and analysis of the main theories, innovative methods, and key technologies of PCC technology, as well as its current status and future development.
Based on this, the team led by Academician Keqiang Li from Tsinghua University conducted a comprehensive review of the technological progress of PCC by consulting relevant global literature in the field, and provided application examples of PCC in typical scenarios under cloud control system and discussed its future development. The article first provides an overview of the core theories and key technologies of PCC, and then, based on typical scenarios such as highways and urban road traffic, systematically conducts a detailed summary and analysis of PCC technology. Finally, the article introduces the innovative architecture and typical applications of PCC technology under cloud control systems, and offers prospects for the future trends and development of PCC technology.
Intelligent and connected vehicles are the core industry of the new round of technological revolution. Both single-vehicle intelligent driving and traditional vehicle networking solutions face limitations such as restricted onboard perception capabilities, insufficient controller computing power, and a lack of cross-domain collaborative driving capabilities, making them hard to handle complex and dynamic road traffic scenarios. As a result, their industrial applications are constrained. Cloud control networked intelligent driving technology, leveraging new generation mobile internet and cloud computing capabilities, establishes a networked cross-domain perception and integrated control system. This integration bridges the gap between intelligent vehicles and smart transportation, leading to a significant enhancement in the overall performance of road traffic and vehicle operation.
Predictive Cruise Control (PCC) is an intelligent control technology based on road traffic information. In comparison to traditional Constant Cruise Control (CC) and Adaptive Cruise Control (ACC), PCC technology offers further enhancements in vehicle safety and fuel efficiency, as well as improvements in traffic efficiency. This is attributed to PCC's ability to adjust vehicle speed and power system proactively based on advance information about future road traffic conditions (as depicted in Figure 1). It demonstrates superior predictive optimal driving characteristics in both urban and highway scenarios.
Extensive research indicates that the PCC control effectiveness is closely related to factors such as the geometric resolution of static road information, the accuracy of dynamic traffic information prediction, power system configuration, prediction range, and step size. However, due to the dynamic and unpredictable nature of traffic environments, as well as the diversity of constraint boundaries and the complexity of solving problems, further improvements in the application effectiveness of PCC technology are constrained. Therefore, the key research area in current PCC technology lies in how to enhance the precision of traffic condition prediction and accelerate the solution of predictive optimal control problems..
Fig. 1 Predictive information available to PCC-equipped vehicles
As mentioned earlier, the PCC system requires substantial support in terms of both static and dynamic road traffic information, as well as real-time computation. There are bottlenecks in the computing power and perception range of the onboard computing platform, which hinders the further industrial development of PCC technology. In recent years, the emergence of the Intelligent and Connected Vehicle Cloud Control System (ICVCCS) has provided a new solution to these challenges. As depicted in Figure 2, ICVCCS aggregates real-time information from the vehicle-road-cloud-network-map. It not only enables cross-domain/long-term perception and cognitive fusion computation but also offers rapid computing services for various intelligent driving applications in the cloud, significantly alleviating the computational burden on the vehicle side. It can be anticipated that future advancements in ICVCCS-enhanced PCC technology, represent a pivotal development direction in the field of intelligent connected vehicle driving control technology. Additionally, it stands as a forefront research direction in the large-scale industrial popularization of cloud-controlled typical applications.
Fig. 2 Cloud control system architecture of intelligent and connected vehicles
This paper conducts a comprehensive investigation of PCC-related methods by reviewing a substantial body of literature both domestically and internationally. The research is conducted in various scenarios, including urban roads and highways. Furthermore, to demonstrate the advantages of PCC technology based on ICVCCS, this paper introduces three typical application studies. Specifically, the article first elucidates the fundamental technical principles of PCC technology at a theoretical level. Subsequently, it extensively surveys and summarizes the current research status in the field of PCC across different scenarios. Finally, it expounds on the key technical issues and future development trends of PCC, particularly focusing on the cloud-controlled PCC technology based on networked vehicles.
The organizational framework of this paper is as follows:
Chapter 2 provides a comprehensive summary of the principles and methods related to PCC technology, including the formulation of optimal problems based on predictive information and commonly used solving methods. Chapter 3 and 4 offer detailed introductions to the current research status of PCC technology in highway and urban road traffic scenarios, respectively. Chapter 5 presents PCC-related innovative technologies and typical applications under the ICVCCS architecture. Chapter 6 outlines the research challenges and future trends of PCC technology. Finally, Chapter 7 concludes the entire paper.
Download full text：
Bolin Gao, Keke Wan, Qien Chen, Zhou Wang, Rui Li, Yu Jiang, Run Mei, Yinghui Luo, Keqiang Li