Demonstration for IMPACT

IMPACT: Interactive Mixed-digital-twin Paradigm for Advanced Cooperative vehicle-infrastructure Testing.

Jianghong Dong1, Chunying Yang1 Mengchi Cai1, Chaoyi Chen1,
Qing Xu1,*, Jianqiang Wang1, Keqiang Li1
1Tsinghua University

Experimental setup


Descriptive Alt Text

Experiment: Human-injected Brake Perturbations

Human operator injects head-vehicle braking disturbance via the MR HMD (Mixed Reality Head-Mounted Display).

More Interactive VICS Testing Experiments

1. Human-triggered Obstacle Spawning

Human-triggered Obstacle Spawning

2. Human-triggered Emergency Obstacle Avoidance: Success

3. Human-triggered Emergency Obstacle Avoidance: Failure

Interactive Testing of Typical VICS Applications

1. Smart Streetlight: Physical Light + Physical Vehicle (Traditional Way)

Virtual Lights with Virtual Vehicles in simulation is another more widely adopted testing approach.

2. Smart Streetlight: Virtual Light + Manual Physical Vehicle (IMPACT's Way)

The vehicle is physical and controlled by a human via a game controller, directly introducing stochastic human behaviors alongside real-world vehicle state update rates and precision, while the streetlights are virtual to avoid energy costs caused by frequent switching.

3. Smart Barrier Gate: Physical Gate + Physical Vehicle (Traditional Way)

Virtual Gates with Virtual Vehicles in simulation is another more widely adopted testing approach.

4. Smart Barrier Gate: Virtual Gate + Manual Physical Vehicle (IMPACT's Way)

Virtual barrier gates and physical vehicles are used to prevent potential collision damage. The virtual vehicle, serving as the digital twin of the physical one, illustrating synchronized operation and real-time cross-platform interaction between the physical and virtual environments.

Abstract

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.

Detailed architecture of IMPACT


Descriptive Alt Text

IMPACT Implementation: I-VIT Experimental Platform


Descriptive Alt Text

I-VIT: Interactive Vehicle-Infrastructure Testbed.