Sufficient testing under corner cases is critical for the long-term operation of vehicle-infrastructure cooperation systems (VICS). However, existing corner-case generation methods are primarily AI-driven, and VICS testing under corner cases is typically limited to simulation. In this paper, we introduce an L5 ''Interactable'' level to the VICS digital twin (VICS-DT) taxonomy, extending beyond the conventional L4 ''Optimizable'' level. We further propose an L5-level VICS testing framework, IMPACT (Interactive Mixed-digital-twin Paradigm for Advanced Cooperative vehicle-infrastructure Testing). By enabling direct human interactions with VICS entities, IMPACT incorporates highly uncertain and unpredictable human behaviors into the testing loop, naturally generating high-quality corner cases that complement AI-based methods. Furthermore, the mixedDT-enabled ''Physical-Virtual Action Interaction'' facilitates safe VICS testing under corner cases, incorporating real-world environments and entities rather than purely in simulation. Finally, we implement IMPACT on the I-VIT (Interactive Vehicle-Infrastructure Testbed), and experiments demonstrate its effectiveness. The experimental videos are available at our project website: https://dongjh20.github.io/IMPACT.
Figures & tables
Figure 1: Schematics for classical DT and mixedDT. (a) In classical DT, virtual entities are typically digital counterparts of the physical ones. (b) In mixedDT, virtual entities exist independently and the physical and virtual entities are integrated into the unified mixed space, where they can coexist and interact with each other.
Level
Capability
Requirement
L1 Visualizable [ 20 , 21 ]
Mirroring the physical via the virtual (Holographic Perception)
Emerging perception data, requiring holographic data governance.
L2 Diagnosable [ 22 , 23 ]
Diagnosing the physical via the virtual (Causal Inference)
Spatio-temporally continuous detection data, requiring causal diagnosis of the road traffic system.
L3 Predictable [ 24 , 25 ]
Predicting the physical via the virtual (Decision Support)
Numerous control strategies with high effect uncertainty, requiring decision support.
L4 Optimizable [ 13 , 26 ]
Optimizing the physical via the virtual (Autonomic Control)
Two-way data interaction, requiring autonomic closed-loop traffic control.
L5 Interactable ∗
Actuating the physical via the virtual (Actionable Interaction)
Bidirectional action-level interaction between physical and virtual entities, requiring direct human intervention in their control.
Note: Levels L1–L4 are adapted from Sun et al. [ 27 ] . Level L5 ( ∗ ) is newly proposed in this study.
Table 1: Digital Twin Levels of VICS
Figure 2: Schematic of I-VIT, IMPACT’s practical implementation. In the mixed testbed, the physical vehicles and infrastructure in the physical testbed (outlined in gold), the virtual vehicles and infrastructure in the virtual testbed (outlined in cyan), the virtual vehicle in the virtual environment of a driving simulator (outlined in red) and human operators in the HMI testbed operate synchronously and interact seamlessly across environments in real time. Consequently, I-VIT is an interactive VICS-DT testing environment, where human operators can directly interact with the physical and virtual entities in I-VIT. The demonstration videos are available at our project website: https://dongjh20.github.io/IMPACT .
Figure 3: System architecture of the IMPACT framework, consisting of the physical, virtual, mixed testing environments and the interactive environment.
Figure 4: Mechanisms for scenario visualization and operational intent processing via HMI devices and their corresponding implementation examples.
Figure 5: Illustration of data interactions within the IMPACT framework and its implementation, the I-VIT platform.
Figure 6: Illustration of the I-VIT architecture and interactions among its four components: physical, virtual, HMI, and mixed testbeds.
Figure 7: Physical and virtual traffic environments, vehicles and facilities. (a) The physical traffic environment, having a size of 9m×5m . (b) The virtual traffic environment, developed based on the Unity game engine. (c) The serial channel-facility mapping table for physical roadside facilities. Blue denotes streetlights and traffic signals, whereas orange denotes barrier gates. (d) Physical and virtual vehicles and roadside facilities.
Component
Setup
Physical sand table
9m×5m , 1:14 with respect to the real-world driving environment
Roadside sensors in the physical testbed
16 identical cameras surround the physical sand table
Roadside facilities in the physical testbed
73 streetlights, 11 traffic signals and 3 barrier gates
Figure 8: Driving view and information board. In (a) and (b), the snapshots of physical and virtual views are presented respectively; see the screens for the specific view. In (c), the real-time speed of the preceding vehicle in the virtual driving view is displayed on the information board on the top of the preceding vehicle.
Table 3: Measurement of Communication Delays in I-VIT (Corresponding to Fig. 5(b) , Unit: ms )
Figure 9: Illustration of the experimental setup.
Figure 10: Snapshots of the experiment. (a) Visualization of the entire platoon projected into the virtual space by Unity. The platoon configuration is presented in Fig. 9 . (b) A human operator controls the last vehicle (Vehicle 8 ) via the InnoSimulation simulator. The experimental videos are available at our project website: https://dongjh20.github.io/IMPACT .
Figure 11: Velocity and distance profiles of all vehicles in the experiments. Two braking disturbances are randomly applied to the head vehicle by the human operator through the MR HMD, which induces speed fluctuations throughout the entire platoon.
In the emerging mixed traffic environments, Connected and Autonomous Vehicles (CAVs) have to interact with surrounding human-driven vehicles (HDVs). This paper introduces MSH-MCCT (Multi-Source Human-in-the-Loop Mixed Cloud Control Testbed), a novel CAV testbed that captures complex interactions between various CAVs and HDVs. Utilizing the Mixed Digital Twin concept, which combines Mixed Reality with Digital Twin, MSH-MCCT integrates physical, virtual, and mixed platforms, along with multi-source control inputs. Bridged by the mixed platform, MSH-MCCT allows human drivers and CAV algorithms to operate both physical and virtual vehicles within multiple fields of view. Particularly, this testbed facilitates the coexistence and real-time interaction of physical and virtual CAVs & HDVs, significantly enhancing the experimental flexibility and scalability. Experiments on vehicle platooning in mixed traffic showcase the potential of MSH-MCCT to conduct CAV testing with multi-source real human drivers in the loop through driving simulators of diverse fidelity. The videos for the experiments are available at our project website: https://dongjh20.github.io/MSH-MCCT.
Jianghong Dong, Chunying Yang, Mengchi Cai +5
School of Vehicle and Mobility, Tsinghua University, Beijing 100084, China. · Department of Civil and Environmental Engineering, University of Michigan, Ann Arbor, MI 48109, USA.
Cooperative autonomous driving in the IoT-to-Edge-to-Cloud continuum requires system-level validation across vehicles, infrastructure sensors, edge-side Dynamic Map services, and in-vehicle autonomous-driving stacks. This paper presents VAST, a V2X/Dynamic Map-aware validation toolchain that connects Scenic, Scenario Simulator v2, AWSIM, Autoware, and SIM-LDM. VAST does not introduce a new search algorithm; instead, it addresses interoperability challenges, including Lanelet2-to-Scenic mapping, ROS 2-based co-simulation through SS2, Dynamic Map object injection into Autoware, and collection of TTC, PET, collision, timeout, and performance measurements. In occluded-intersection scenarios, Lanelet2-compatible constrained sampling increases the edge-case discovery rate from 40.0% to 80.0% and reduces the average time per discovered edge case from 259.7 s to 110.4 s. Under the same generated scenario distribution, Dynamic Map availability reduces the collision rate from 78.0% to 40.0% and increases non-collision outcomes from 22.0% to 60.0%, with statistically significant TTC/PET shifts. A throughput study with 1-16 NPCs shows that sampling remains below 0.1 s, whereas AWSIM/Autoware execution and restart overhead dominate runtime. These results position VAST as a practical validation infrastructure for cooperative autonomous-driving CPSs.
Shunsuke Ito, Takuya Azumi
Graduate School of Science and Engineering Saitama University · Academic Association (Graduate School of Science and Engineering) Saitama University
Comprehensive and efficient validation of connected and automated vehicles (CAVs) is critical prior to real-world deployment. While simulation-based testing offers scalability, existing approaches often lack seamless integration with real vehicles and field data, limiting their fidelity in capturing dynamic, real-world interactions. To bridge this gap, this paper proposes a novel real-time hybrid digital twin platform. Its core innovation lies in the tight coupling of a high-fidelity CARLA-SUMO co-simulation with a physical test site and vehicle via a low-latency Vehicle-to-Everything (V2X) communication link. A custom-developed middleware serves as the critical bridge, synchronizing a real CAV's kinematic state as a shadow vehicle in the simulation and translating virtual control commands into chassis-actuating Controller Area Network (CAN) messages for closed-loop control. Detailed implementation includes using photogrammetry for full-scale asset reconstruction and a cloud-edge collaborative architecture for scalable, multi-user operation. Experimental results demonstrate stable synchronization and effective closed-loop control with low latency, confirming the platform's practicality for multi-scenario CAV verification.
Kanglong Quan, Zhebing Xia, Linfeng Jiang +4
School of Advanced Technology Xi’an Jiaotong-Liverpool University, Jiangsu, China