Abstract: Parallel transportation serves as a holistic paradigm for achieving intelligent traffic management and control, focusing on addressing the complexity of human and social factors. Recently, the emergence and development of foundational models (FMs) have ushered in a new era for the realization of parallel transportation. However, the inherent issues of “hallucinations,” outdated knowledge, and the “black-box” nature of FMs render their generated decisions unreliable and untrustworthy. To address these issues, we propose a TransRAG framework for parallel transportation based on retrieval-augmented generation and chain-of-thought (CoT) prompting. TransRAG is composed of three interacting layers, storage, management, and execution, which work together to deliver personalized and diverse traffic services to users. The external knowledge from the storage layer is incorporated to augment the FM in management layers for computational experiments. The real–virtual interaction between artificial and actual transportation systems is used to continuously optimize the decisions from the management layer. Therefore, TransRAG can incrementally update knowledge and adjust strategies to adapt to the evolving and dynamic traffic environment. Additionally, the integration of blockchain, smart contracts, and caching into TransRAG is expected to address a range of challenges, such as single point of failure, potential privacy breaches, and delays in data access, thereby advancing the transition to “6S” Transportation 5.0.

基于检索增强生成的安全可信平行交通框架

[1]Chen JW, Lin HY, Han XP, et al., 2024. Benchmarking large language models in retrieval-augmented generation. Proc 38^th AAAI Conf on Artificial Intelligence, p.17754-17762.

[2]Dai XY, Guo C, Tang Y, et al., 2024. VistaRAG: toward safe and trustworthy autonomous driving through retrieval-augmented generation. IEEE Trans Intell Veh, 9(4):4579-4582.

[3]Feng GH, Zhang BH, Gu YT, et al., 2024. Towards revealing the mystery behind chain of thought: a theoretical perspective. Proc 37^th Int Conf on Neural Information Processing Systems, p.70757-70798.

[4]Gan L, Chu WB, Li GF, et al., 2024. Large models for intelligent transportation systems and autonomous vehicles: a survey. Adv Eng Inform, 62:102786.

[5]Jin K, Wi J, Lee E, et al., 2021. TrafficBERT: pre-trained model with large-scale data for long-range traffic flow forecasting. Expert Syst Appl, 186:115738.

[6]Lai GP, Liu MK, Wang FY, et al., 2001. Web caching: architectures and performance evaluation survey. Proc IEEE Int Conf on Systems, Man and Cybernetics. e-Systems and e-Man for Cybernetics in Cyberspace, p.3039-3044.

[7]Lewis P, Perez E, Piktus A, et al., 2020. Retrieval-augmented generation for knowledge-intensive NLP tasks. Proc 34^th Int Conf on Neural Information Processing Systems, p.9459-9474.

[8]Li JJ, Qin R, Guan ST, et al., 2024. Blockchain intelligence: intelligent blockchains for Web 3.0 and beyond. IEEE Trans Syst Man Cybern Syst, 54(11):6633-6642.

[9]Ouyang LW, Zhang WW, Wang FY, 2022. Intelligent contracts: making smart contracts smart for blockchain intelligence. Comput Electr Eng, 104:108421.

[10]Sanderson K, 2023. GPT-4 is here: what scientists think. Nature, 615(7954):773.

[11]Wang FY, 2008. Toward a revolution in transportation operations: AI for complex systems. IEEE Intell Syst, 23(6):8-13.

[12]Wang FY, 2010. Parallel control and management for intelligent transportation systems: concepts, architectures, and applications. IEEE Trans Intell Transp Syst, 11(3):630-638.

[13]Wang FY, Li JJ, Qin R, et al., 2023a. ChatGPT for computational social systems: from conversational applications to human-oriented operating systems. IEEE Trans Comput Soc Syst, 10(2):414-425.

[14]Wang FY, Lin YL, Ioannou PA, et al., 2023b. Transportation 5.0: the DAO to safe, secure, and sustainable intelligent transportation systems. IEEE Trans Intell Transp Syst, 24(10):10262-10278.

[15]Wang FY, Miao QH, Li LX, et al., 2024. When does Sora show: the beginning of TAO to imaginative intelligence and scenarios engineering. IEEE/CAA J Autom Sin, 11(4):809-815.

[16]Wang S, Yuan Y, Wang X, et al., 2018. An overview of smart contract: architecture, applications, and future trends. IEEE Intelligent Vehicles Symp, p.108-113.

[17]Wang X, Yang J, Han JP, et al., 2022. Metaverses and DeMetaverses: from digital twins in CPS to parallel intelligence in CPSS. IEEE Intell Syst, 37(4):97-102.

[18]Wang X, Wang YT, Yang J, et al., 2024. The survey on multi-source data fusion in cyber-physical-social systems: foundational infrastructure for industrial metaverses and Industries 5.0. Inform Fus, 107:102321.

[19]Wei J, Wang XZ, Schuurmans D, et al., 2022. Chain-of-thought prompting elicits reasoning in large language models. Proc 36^th Int Conf on Neural Information Processing Systems, p.24824-24837.

[20]Yang J, Wang X, Tian YL, et al., 2023a. Parallel intelligence in CPSSs: being, becoming, and believing. IEEE Intell Syst, 38(6):75-80.

[21]Yang J, Ni QH, Luo GY, et al., 2023b. A trustworthy Internet of Vehicles: the DAO to safe, secure, and collaborative autonomous driving. IEEE Trans Intell Veh, 8(12):4678-4681.

[22]Zeng D, Wang FY, Liu MK, 2004. Efficient web content delivery using proxy caching techniques. IEEE Trans Syst Man Cybern C Appl Rev, 34(3):270-280.

[23]Zhang KP, Zhou F, Wu L, et al., 2024. Semantic understanding and prompt engineering for large-scale traffic data imputation. Inform Fus, 102:102038.

[24]Zhou J, Ke P, Qiu XP, et al., 2023. ChatGPT: potential, prospects, and limitations. Front Inform Technol Electron Eng, early access.

[25]Zhu FH, Lv YS, Chen YY, et al., 2020. Parallel transportation systems: toward IoT-enabled smart urban traffic control and management. IEEE Trans Intell Transp Syst, 21(10):4063-4071.