Socially compliant automated vehicles: new conceptual framework paves the way for safer mixed-traffic environments
Tsinghua University Press
image:
The proposed conceptual framework integrates sensing, socially‑compliant decision‑making, safety constraints, bidirectional behavioral adaptation, and spatial‑temporal memory to guide future SCAV development.
view moreCredit: Communications in Transportation Research
Automated vehicles (AVs) hold promise for revolutionizing transportation by improving road safety, traffic efficiency, and overall mobility. Despite the steady advancement in high-level AVs in recent years, the transition to full automation entails a period of mixed traffic, where AVs of varying automation levels coexist with human-driven vehicles (HDVs). Making AVs socially compliant and understood by human drivers is expected to improve the safety and efficiency of mixed traffic. Thus, ensuring AVs’ compatibility with HDVs and social acceptance is crucial for their successful and seamless integration into mixed traffic.
Researchers at Delft University of Technology (Netherlands) and RWTH Aachen University (Germany) carried out a study with a comprehensive scoping review to assess the current state of the art in developing socially compliant automated vehicles (SCAVs), identifying key concepts, methodological approaches, and research gaps. They conducted an informal expert interview to discuss the literature review results and identify critical research challenges and expectations towards SCAVs. Based on the scoping review and expert interview input, they designed a conceptual framework for the development of SCAVs and evaluated it using an online survey targeting researchers, technicians, policymakers, and other relevant professionals worldwide. The framework outlines the key capability elements necessary for SCAVs and incorporates crucial considerations across technical, social, and cultural dimensions, effectively bridging theoretical insights with practical applications to achieve socially compliant automation. The conceptual framework provides actionable insights into developing and embedding social compliance in AV systems, enabling scalable and context-sensitive deployment. It can also foster collaboration among academia, industry, and policymakers, ensuring technical innovation aligns with societal needs and regulatory standards, accelerating the path toward SCAV and further towards safe and socially inclusive automated mobility solutions.
The study was published on 8 September 2025, in Communications in Transportation Research.
“We conducted the first comprehensive scoping review to identify key concepts, methodologies, and research gaps in the emerging field of socially compliant automated driving in mixed traffic,” says Yongqi Dong, lead author and researcher at Delft University of Technology and RWTH Aachen University. “By combining this review with expert interviews and a worldwide survey, we have distilled core requirements for AVs to coexist seamlessly and predictably with human drivers through a conceptual framework.” (The conceptual framework is shown in the figure.)
Five methodological pillars
In reviewing 68 pivotal studies on SCAVs, the team clustered existing approaches into five main categories:
- Imitation learning to clone human social driving norms
- Reinforcement learning with utility-based models
- Model‑based (e.g., game theory, social‑force models, and driving risk field models) generation of human-like behaviors
- Socially‑aware trajectory prediction with social factors and machine learning
- Optimization of social driving parameters balancing safety, comfort, and courtesy
Expert insights reveal critical gaps
Through informal interviews with ten AV experts across academia, industry, and government, the authors uncovered key limitations in today’s AVs:
- Excessive conservatism, leading to inefficient traffic flow
- Poor interpretation of implicit human cues, from hand gestures to assertive lane changes
- Inflexibility to diverse driving cultures and styles
- Lack of bidirectional adaptation, where AVs and human drivers dynamically adjust to each other
A novel conceptual framework
To address these gaps, the author team proposes a novel conceptual framework for SCAV (see Image) comprising:
- Sensing & perception module, fusing multi-modal sensor data
- Socially‑compliant decision‑making module, embedding social components (including culture, norms, and cues), different driving styles (e.g., aggressive, cautious, pro-social), and bidirectional behavioral adaptation mechanisms
- Safety constraints module, a real‑time safeguard layer
- Utility trade‑off mechanisms, balancing individual vehicles’ benefits with network‑level traffic performance
- Bidirectional behavioral adaptation, enabling AVs to learn from and respond to human drivers’ adaptations (there will be bidirectional iterative adaptations between AVs and human drivers)
- Spatial‑temporal memory module, continuously updating AV models with past interaction data to facilitate the long- and short-term updating of knowledge and driving rules and contribute to the implementation of bidirectional behavioral adaptation.
Global survey validation
An online questionnaire gathered insights from 90 experts across 29 countries. Respondents overwhelmingly endorsed the framework’s key modules, particularly the need for mutual bidirectional adaptation between AVs and human drivers, and highlighted priorities such as anticipation capability, multi-objective optimization, and spatial-temporal memory buffer integration.
“Our framework lays a robust foundation for both academia and industry,” concludes Bart van Arem, co‑author from Delft University of Technology. “By integrating social dynamics into AV design, we can pave the way toward mixed‑traffic environments that are safer, more efficient, socially beneficial, and widely accepted.”
This research marks a significant step toward realizing the promise of automated vehicles in everyday traffic, ensuring they not only drive safely but also drive in harmony with human behavior and societal expectations.
The above research is published in Communications in Transportation Research (COMMTR), which is a fully open access journal co-published by Tsinghua University Press and Elsevier. COMMTR publishes peer-reviewed high-quality research representing important advances of significance to emerging transport systems. COMMTR is also among the first transportation journals to make the Replication Package mandatory to facilitate researchers, practitioners, and the general public in understanding and advancing existing knowledge. At its discretion, Tsinghua University Press will pay the open access fee for all published papers in 2025.
Journal
Communications in Transportation Research
Article Title
Towards developing socially compliant automated vehicles: Advances, expert insights, and a conceptual framework
Can generative AI improve vehicle trajectory prediction in car-following scenarios?
Tsinghua University Press
image:
The proposed FollowGen framework consists of four main modules: historical feature encoding, noise scaling strategy and noise addition, car-following vehicular interaction modeling, and condition guided denoising.
view moreCredit: Communications in Transportation Research
To answer this question, researchers from the University of Wisconsin–Madison, Tongji University, and collaborators developed FollowGen, a conditional diffusion model that integrates historical motion features and inter-vehicle interactions to generate safer and more reliable trajectory predictions for autonomous driving.
They published their study on 16 October 2025, in Communications in Transportation Research.
“We introduce a scaled noise conditioning mechanism that embeds historical motion features into the forward diffusion process. This allows the model to account for motion-aware uncertainties from the beginning and generate trajectories that better align with real driving behaviors,” says Junwei You, a Ph.D. candidate studying autonomous driving and intelligent transportation.
Different performance across diverse driving scenarios
In the study, the research group evaluated FollowGen using multiple real-world datasets, covering both urban and highway environments. They examined car-following situations such as human-driven vehicles following each other, autonomous vehicles following humans, and humans following autonomous vehicles.
“This design enables the model to explicitly capture the interaction between leading and following vehicles. As a result, FollowGen delivers consistent improvements over strong baselines, particularly in terms of final displacement accuracy and reliability,” explains Haotian Shi, Associate Professor at Tongji University.
Visualization results show how initially chaotic scaled noise progressively evolves into accurate trajectory predictions, demonstrating the effectiveness of combining diffusion models with car-following dynamics.
Significant implications for autonomous driving
The findings suggest that incorporating generative AI into trajectory prediction can enhance the safety and robustness of autonomous driving systems. Beyond real-time planning, the model also offers potential applications in large-scale traffic simulation and intelligent transportation system design.
“Our research demonstrates that diffusion-based models are not only powerful for uncertainty modeling but can also be adapted to the unique requirements of vehicle interactions on the road. This represents an important step toward safer and more reliable autonomous vehicles,” says Sikai Chen, Assistant Professor at the University of Wisconsin-Madison.
The above research is published in Communications in Transportation Research (COMMTR), which is a fully open access journal co-published by Tsinghua University Press and Elsevier. COMMTR publishes peer-reviewed high-quality research representing important advances of significance to emerging transport systems. COMMTR is also among the first transportation journals to make the Replication Package mandatory to facilitate researchers, practitioners, and the general public in understanding and advancing existing knowledge. At its discretion, Tsinghua University Press will pay the open access fee for all published papers in 2025.
Journal
Communications in Transportation Research
Article Title
Follow Gen: A scaled noise conditional diffusion model for car-following trajectory prediction
Can traffic accident reports aid visual accident anticipation?
Tsinghua University Press
image:
The traditional approach (top) relies on object detection, depth estimation, and optical flow processed through a Graph Convolutional Network for accident anticipation. In contrast, our proposed method (bottom) integrates domain knowledge and a Large Language Model (GPT-4o) to enhance interpretability and provide more context-aware feedback.
view moreCredit: Communications in Transportation Research
To address this question, a research team from the University of Macau designs a dual-branch vision-language framework that incorporates domain knowledge as a mediating factor to evaluate the role of textual information in visual reasoning tasks.
They published their study on 14 October 2025, in Communications in Transportation Research.
“In the field of accident anticipation, textual accident reports and traffic accident videos are traditionally studied in isolation. However, we argue that there exists an intrinsic relationship between the two modalities. Benefiting from recent advances in vision-language models, we are able to explore this relationship and assess the contribution of accident reports to accident anticipation tasks.”, says Yanchen Guan, a researcher at the Department of Civil Engineering at University of Macau.
Domain-Enhanced Dual-Branch Model
Real-time traffic accident prediction is a critical component of safety systems in autonomous vehicles. By anticipating potential accidents, autonomous systems can identify and respond to imminent hazards in a timely manner, thereby reducing the risk of injury and property damage. However, achieving high-performance and interpretable accident prediction under constrained computational resources remains a significant challenge.
In the study, the research team proposes a deep learning architecture that integrates visual and textual features through domain knowledge, aiming to develop a lightweight, high-accuracy, and interpretable real-time traffic accident prediction system. The framework extracts accident-related factors from accident reports as prior knowledge to assist scene-level accident anticipation. Finally, it leverages this prior knowledge to guide a large language model in generating contextually appropriate driving suggestions and archiving predicted traffic accidents.
Domain knowledge is Helpful for Accident Anticipation
The proposed model is evaluated on three real-world datasets—DAD, CCD, and A3D—and achieves strong performance across all benchmarks. The results demonstrate that incorporating domain knowledge as a mediating layer to decompose traffic scenes into contributing factors not only assists the model in making accurate predictions, but also guides the attention of large language models toward accident-inducing elements, thereby enabling the generation of targeted driving suggestions.
This study provides valuable insights into the domain of traffic accident prediction and presents a high-accuracy, computationally efficient inference framework. It reveals the underlying connections between textual and visual data, introducing a new research direction that integrates multimodal information for interpretable and efficient accident anticipation. These findings have practical implications for enhancing the safety systems of autonomous vehicles.
Future work can further explore the latent correspondence between accident reports and video data, transforming large-scale textual records into richly annotated visual data to support autonomous driving model training. In addition to binary accident prediction, future studies may also refine scene understanding and deliver context-aware, scenario-specific driving recommendations.
The above research is published in Communications in Transportation Research (COMMTR), which is a fully open access journal co-published by Tsinghua University Press and Elsevier. COMMTR publishes peer-reviewed high-quality research representing important advances of significance to emerging transport systems. COMMTR is also among the first transportation journals to make the Replication Package mandatory to facilitate researchers, practitioners, and the general public in understanding and advancing existing knowledge. At its discretion, Tsinghua University Press will pay the open access fee for all published papers in 2025.
About Communications in Transportation Research
Journal
Communications in Transportation Research
Article Title
Domain-Enhanced Dual-Branch Model for Efficient and Interpretable Accident Anticipation
How can Modular Autonomous Vehicles achieve safe docking and undocking on the road?
image:
The results show the travel distance, speed, headway, and perceived infrared intensity of two MAVs over time. The following vehicle adjusts its motion based on noisy perception and control barrier function constraints, enabling safe and efficient docking and undocking under uncertainty.
view moreCredit: Communications in Transportation Research
Modular Autonomous Vehicles (MAVs) can dynamically connect and disconnect en route, enabling flexible and efficient passenger and freight transportation. However, ensuring safety during the docking and undocking process remains a major challenge due to close-proximity operation, perception noise, and system uncertainties. To address these challenges, researchers at the University of Wisconsin–Madison, USA, proposed a Safety Assurance Adaptive Model Predictive Control (SAAMPC) framework. The method combines adaptive control and safety barrier functions to handle real-time disturbances and ensure robust operation. The effectiveness of the approach was validated through both simulations and reduced-scale physical experiments using cars.
They published their study on 3 September 2025, in Communications in Transportation Research.
Modular Autonomous Vehicles (MAVs) have the potential to transform transportation by allowing multiple vehicle units to connect and disconnect dynamically. However, how to safely and reliably control such docking and undocking operations in uncertain environments remains an open challenge.
“Our work is to move beyond theoretical modeling and simulation to implement and verify MAV docking control in a physical testbed,” says Chengyuan Ma, a postdoctoral researcher at the University of Wisconsin–Madison. “We developed a Safety Assurance Adaptive Model Predictive Control (SAAMPC) framework that combines model predictive control with adaptive parameter tuning and safety assurance using control barrier functions. This allows the MAVs to maintain safe operation despite perception noise, control uncertainty, and external disturbances.”
Adaptive safety constraints ensure robust control under uncertainty
“One of the key features of our framework is the use of adaptive Control Barrier Functions (CBFs),” says Chengyuan Ma. “Instead of setting fixed safety limits, our method dynamically adjusts the CBF constraints in real time based on perception reliability and control stability. This ensures the following vehicle does not over-accelerate during docking, even under disturbances or sensor noise.”
Validated through both simulation and reduced-scale physical testing
“To ensure practical applicability, we validated our method in both simulation and real-world conditions,” Chengyuan Ma explains. “We used Simulink to test the SAAMPC framework under various disturbances, and conducted physical experiments with robot vehicles on a circular track. The experimental results confirmed that our method works reliably even in the presence of real-world noise, uncertainty, and limited sensing.”
The results demonstrate that our SAAMPC framework enables smooth, safe, and robust docking and undocking operations. Our work lays a technical foundation for the future deployment of modular vehicle systems in real-world transportation networks
The above research is published in Communications in Transportation Research (COMMTR), which is a fully open access journal co-published by Tsinghua University Press and Elsevier. COMMTR publishes peer-reviewed high-quality research representing important advances of significance to emerging transport systems. COMMTR is also among the first transportation journals to make the Replication Package mandatory to facilitate researchers, practitioners, and the general public in understanding and advancing existing knowledge. At its discretion, Tsinghua University Press will pay the open access fee for all published papers in 2025.
Journal
Communications in Transportation Research
Article Title
Safety Assurance Adaptive Control for Modular Autonomous Vehicles
KIDL: A knowledge-informed deep learning paradigm for generalizable and stability-optimized car-following models
image:
The knowledge-informed deep learning (KIDL) paradigm, with the blue section representing the LLM workflow (teacher demonstration), the orange section representing the distillation pipeline of KIDL (student learning), and the green section representing the stability optimization process.
view moreCredit: Communications in Transportation Research
As autonomous vehicles (AVs) integrate into real-world traffic, a key challenge remains: how to ensure their car-following behavior is both human-like and traffic-stable. Traditional car-following models (CFMs), whether physics-based or data-driven, often fail to generalize across diverse driving scenarios or maintain traffic flow stability.
This research, published on 20 September 2025, in Communications in Transportation Research, is a collaborative effort among Xi’an Jiaotong-Liverpool University, Monash University, University of Liverpool, and City University of Hong Kong, integrating expertise in transportation engineering, machine learning, and intelligent systems.
The study introduces KIDL—Knowledge-Informed Deep Learning—a groundbreaking framework that distills expert knowledge from Large Language Models (LLMs) into a lightweight and stability-optimized neural network for AV control. Unlike prior models limited by dataset biases, KIDL captures generalized driving principles by querying LLMs for thousands of diverse traffic scenarios and refining predictions through a structured knowledge distillation process.
KIDL’s innovation lies not only in its strong behavioral generalization but also in its theoretical guarantees: by incorporating local and string stability constraints into training, it ensures smooth and safe vehicle interactions in mixed traffic. Experiments on real-world datasets (NGSIM, HighD) confirm KIDL outperforms physics-based, deep learning, and hybrid models in both accuracy and safety—with zero collisions and superior generalization.
KIDL represents a new paradigm in autonomous vehicle design, blending human-like reasoning with formal stability control—scalable, interpretable, and ready for real-world deployment.
Car-following models (CFMs) are microscopic traffic models that capture longitudinal interactions between leading and following vehicles. Most CFMs follow a model-centric design and are calibrated or trained on specific datasets. While this yields high accuracy within seen scenarios, performance often degrades under unseen conditions due to the out-of-distribution generalization problem. Individual datasets rarely capture the full range of real-world variability, limiting model robustness. Although data-centric approaches that focus on collecting broader datasets can improve generalization, they are costly and difficult to scale. At the same time, ensuring traffic flow stability has become increasingly important for the deployment of CFMs in AV-integrated traffic systems, where safety and efficiency are critical priorities. This highlights the need for a new modeling paradigm that simultaneously addresses the limitations of generalization and the rising demand for stability in mixed traffic environments. To this end, we propose a Knowledge-Informed Deep Learning (KIDL) paradigm that jointly enhances behavioral generalization and traffic flow stability.
KIDL improves generalization by leveraging large language models (LLMs) as knowledge sources that encapsulate high-level car-following behaviors learned from diverse textual data, including driving rules, safety guidelines, and human reasoning. This enables KIDL to capture generalizable principles that extend beyond the scope of any single traffic dataset. Through knowledge distillation, insights from LLMs (as teachers) are transferred to lightweight neural networks (as students), forming a compact and efficient representation. Rather than employing LLMs as end-to-end models, KIDL adopts a distillation-based approach with three key advantages.
The first advantage is computational efficiency. LLMs generate linguistic responses sequentially and require substantial memory and processing resources, making them unsuitable for real-time applications. In contrast, KIDL produces single-step numerical predictions with significantly fewer parameters, enabling real-time inference at a fraction of the computational cost. The second advantage is the prediction reliability. LLMs may produce inaccurate or unfaithful content, which poses serious risks in safety-critical contexts. KIDL reduces this risk by applying self-consistency with majority voting during knowledge extraction, improving reliability and minimizing the likelihood of erroneous behavior.
The third advantage is theoretical tractability. The black-box nature, complex architectures, and dependence on natural language inputs and outputs make LLMs difficult to interpret and unsuitable for formal analysis, limiting their applicability in stability studies such as local and string stability. By distilling knowledge into a simplified surrogate model with numerical inputs and outputs, KIDL enables interpretable and analytically tractable stability analysis.
This property further allows KIDL to incorporate physically grounded stability constraints directly into the training objective, ensuring compliance with both local and string stability conditions. As a result, the model suppresses disturbance amplification and promotes smooth traffic flow.
By integrating behavioral fidelity with stability optimization, KIDL provides a scalable and robust solution for deployment in mixed traffic environments. This combination of generalizable behavior modeling and explicit stability assurance addresses a critical gap between human driver emulation and control-oriented AV deployment. To the best of our knowledge, KIDL is among the first frameworks to systematically achieve both objectives within a unified paradigm by distilling car-following knowledge from LLMs into a stability-aware neural architecture.
The above research is published in Communications in Transportation Research (COMMTR), which is a fully open access journal co-published by Tsinghua University Press and Elsevier. COMMTR publishes peer-reviewed high-quality research representing important advances of significance to emerging transport systems. COMMTR is also among the first transportation journals to make the Replication Package mandatory to facilitate researchers, practitioners, and the general public in understanding and advancing existing knowledge. At its discretion, Tsinghua University Press will pay the open access fee for all published papers in 2025.
Journal
Communications in Transportation Research
Article Title
A knowledge-informed deep learning paradigm for generaliz-able and stability-optimized car-following models
Can combined virtual-real testing speed up autonomous vehicle testing? Findings from AEB field experiments
Tsinghua University Press
image:
Schematic overview of the CVRT system, illustrating its three core modules: the Simulation Environment Module for generating high-fidelity virtual scenarios and traffic flows; the Data Process Module for synchronizing and integrating sensor data from both virtual and real sources; and the Autonomous Vehicle Module where data fusion, decision-making, and path planning are performed. Inter-module communication is enabled via ROS messaging, ensuring seamless coordination between simulation, data processing, and real-world autonomous vehicle operations.
view moreCredit: Communications in Transportation Research
Researchers at Chang’an University have developed a novel combined virtual-real testing (CVRT) platform for validating autonomous vehicles. This innovative approach utilizes digital twin technology to simulate realistic scenarios and conduct parallel AEB tests across various conditions. The results indicate that CVRT closely replicates real-world performance while significantly reducing test time by up to 70%. This breakthrough offers a safer, more efficient method for validating autonomous systems, with implications for scalable testing and regulation in the autonomous vehicle industry.
They published their study on 16 October 2025, in Communications in Transportation Research.
“To overcome the cost and inefficiency of traditional field experiments, we developed a combined virtual-real testing (CVRT) platform based on digital twin technology,” explains Ying Gao, assistant professor and one of the study’s authors. “We systematically compared CVRT with real-world tests for autonomous emergency braking (AEB), using rigorous metrics to ensure the method’s reliability.”
Reliable performance across scenarios and conditions
The team conducted parallel AEB experiments in four challenging scenarios, covering both car-to-car and car-to-pedestrian hazards, as specified by C-NCAP 2024. Each test, which ran at multiple speeds and repeated more than 15 times, combined high-fidelity virtual scenarios with real-world vehicles, sensors, and conditions.
Quantitative results showed that the CVRT platform can closely replicate the speed, trajectory, and acceleration patterns observed in traditional proving ground tests. In particular, the Fréchet distance, a standard metric for measuring similarity between two time series, demonstrated that discrepancies between CVRT and real-world trials were negligible and not statistically significant. “Trigger times for emergency braking were nearly identical, and sensor fusion in the CVRT system proved robust across all tested scenarios,” says Meng Zhang, lead author and PhD candidate.
Efficiency, flexibility, and safer testing
While high fidelity is crucial, CVRT’s biggest advantage may be efficiency. The study found that scenario preparation and test time for a single experiment could be reduced by 40%–70% compared with physical-only experiments, especially as scenarios grow more complex. Automated scenario resets and virtual scene injection accelerate testing while minimizing risk to vehicles and people.
This approach not only expands the variety of hazardous scenarios that can be safely tested, but also reduces costs and logistical hurdles. “With CVRT, we can rapidly iterate, expand scenario coverage, and focus on rare but critical corner cases,” says co-author Jiatong Xu. “It’s a path towards scalable, repeatable, and safer validation of AVs.”
A foundation for the future of AV development
Importantly, the researchers also confirmed that communication delays inherent to the virtual-real integration remained well below the threshold that would impact system safety or performance, a crucial finding for real-world deployment.
“CVRT is poised to redefine how the automotive industry validates autonomy. Our results demonstrate that CVRT can provide the reliability of proving-ground tests, but with much greater speed and flexibility,” explains Professor Zhigang Xu, the project’s supervisor.
While the current study focused on standard AEB scenarios and controlled conditions, the team plans to extend the framework to more diverse driving environments in future work. The open-source test data can be accessed at ETS-Data.
The above research is published in Communications in Transportation Research (COMMTR), which is a fully open access journal co-published by Tsinghua University Press and Elsevier. COMMTR publishes peer-reviewed high-quality research representing important advances of significance to emerging transport systems. COMMTR is also among the first transportation journals to make the Replication Package mandatory to facilitate researchers, practitioners, and the general public in understanding and advancing existing knowledge. At its discretion, Tsinghua University Press will pay the open access fee for all published papers in 2025.
About Communications in Transportation Research
Communications in Transportation Research was launched in 2021, with academic support provided by Tsinghua University and China Intelligent Transportation Systems Association. The Editors-in-Chief are Professor Xiaobo Qu, a member of the Academia Europaea from Tsinghua University and Professor Shuai’an Wang from Hong Kong Polytechnic University. The journal mainly publishes high-quality, original research and review articles that are of significant importance to emerging transportation systems, aiming to serve as an international platform for showcasing and exchanging innovative achievements in transportation and related fields, fostering academic exchange and development between China and the global community.
It has been indexed in SCIE, SSCI, Ei Compendex, Scopus, CSTPCD, CSCD, OAJ, DOAJ, TRID and other databases. It was selected as Q1 Top Journal in the Engineering and Technology category of the Chinese Academy of Sciences (CAS) Journal Ranking List. In 2022, it was selected as a High-Starting-Point new journal project of the “China Science and Technology Journal Excellence Action Plan”. In 2024, it was selected as the Support the Development Project of “High-Level International Scientific and Technological Journals”. The same year, it was also chosen as an English Journal Tier Project of the “China Science and Technology Journal Excellence Action Plan Phase Ⅱ”. In 2024, it received the first impact factor (2023 IF) of 12.5, ranking Top1 (1/58, Q1) among all journals in “TRANSPORTATION” category. In 2025, its 2024 IF was announced as 14.5, maintaining the Top 1 position (1/61, Q1) in the same category. Tsinghua University Press will cover the open access fee for all published papers in 2025.
Journal
Communications in Transportation Research
Article Title
Can combined virtual-real testing speed up autonomous vehicle testing? Findings from AEB field experiments
No comments:
Post a Comment