Revolutionizing Traffic Analysis: Fujitsu and Carnegie Mellon’s ai-Powered Social Digital Twin
In a noteworthy advancement, Fujitsu Limited and Carnegie Mellon University have announced the development of an ai-driven Social Digital Twin technology. This innovative collaboration, initiated in February 2022, aims to transform the field of traffic analysis and accident prevention through high-precision visualization and analysis of dynamic 3D traffic scenarios.
Joint Research Initiative: Creating a Digital Twin of Real-World 3D Spaces
The objective of this research was to develop a Social Digital Twin capable of dynamically replicating complex interactions within 3D spaces. The fruit of their labor is an ai-powered system that can convert 2D scene images captured by a monocular RGB camera into detailed digitalized 3D formats.
The system consists of two core technologies:
1. 3D Occupancy Estimation Technology: Utilizing deep learning networks, this technology distinguishes objects within 3D spaces from monocular RGB camera images and represents them as Voxels in 3D space. This enables accurate 3D shape estimation of areas not visible in the input image.
2. 3D Projection: Once objects have been estimated, this technology builds a 3D digital twin based on the output of the Occupancy Estimation Technology. Incorporating human behavior analysis know-how, it ensures movements are consistent with the real world and enables precise position estimation even when parts of objects are obstructed.
Privacy and Data Security
The technology addresses privacy concerns by automatically anonymizing faces and license plates, ensuring responsible ai use. Fujitsu and Carnegie Mellon plan to commercialize this technology by the fiscal year 2025, expanding its application beyond transportation to encompass smart cities and traffic safety.
Field Trials in Pittsburgh: Validating the Technology’s Potential
With promising capabilities, field trials commenced in Pittsburgh, USA, on February 22, 2024. A monocular RGB camera was installed at Carnegie Mellon University to analyze crowd and traffic conditions, identifying potential accidents. Reproducing this data on a Social Digital Twin allowed for thorough examination of the technology’s effectiveness in real–world scenarios.
Assistant Research Professor, Prof. László A. Jeni, expressed delight at this collaborative achievement. He emphasized the commitment to advancing research on cutting-edge technologies and the pivotal role played by both Fujitsu’s team and Carnegie Mellon University experts.
Daiki Masumoto, Fellow and Head of the Converging Technologies Laboratory at Fujitsu Research, highlighted the technology’s alignment with Fujitsu’s broader mission to make the world more sustainable through innovation. He expressed excitement about the significant step towards achieving their goals through collaboration with Carnegie Mellon University.
Implications of the ai-Powered Social Digital Twin Technology
As the field trials progress, the impact of this groundbreaking technology on traffic analysis and urban planning is becoming increasingly evident. With potential applications in safety, optimizing traffic flow, and contributing to the development of smart cities, this innovation holds substantial promise for enhancing our approach to urban planning, transportation, and societal challenges in the years to come. The question remains: How will this revolutionary technology reshape our cities and societies?
