Cloud-Based Mobility Digital Twin for Scalable Data Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional mobility system frameworks rely heavily on onboard storage and computing, which limits scalability, manageability, and sharing of data, and do not effectively leverage real-time and historical data for predictive analytics and actuation across human, vehicular, and traffic entities.
Innovation Solution
A cloud-based mobility digital twin (MDT) framework that gathers data from physical objects, conforms it to a schema, and transmits it to a cloud-based digital space for processing, enabling actuation instructions to be sent back for operation, using a communications layer and microservices for simulation, machine learning, and prediction across human, vehicular, and traffic digital twins.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If onboard storage and computing are used for mobility data processing, then data processing can be performed locally, but scalability and manageability are limited
Solution Approach 1:
The patent introduces a cloud-based digital twin platform as an intermediary between physical mobility entities and data processing systems. The digital twin serves as a virtual mediator that receives data from sensors and V2X communications, performs comprehensive processing including machine learning and simulation, then sends actuation instructions back to physical entities. This intermediary approach enables scalable cloud-based processing while maintaining local responsiveness through the digital twin-physical entity connection.
2Speed
If onboard computing is used, then real-time processing is possible, but manageability and data sharing effectiveness are reduced
Solution Approach 1:
The patent segments the data processing architecture into multiple components: local sensing units that collect data, a communications layer that transmits data, a cloud-based digital twin platform that performs intensive processing including machine learning and simulation, and an actuation layer that implements decisions. This segmentation allows real-time data collection at the edge while complex processing occurs in the manageable cloud environment, with the digital twin serving as a virtual replica that bridges both environments.
3Adaptability or versatility
If traditional mobility frameworks are used, then existing systems can operate, but predictive analytics and actuation across human, vehicular, and traffic entities are not effectively leveraged
Solution Approach 1:
The patent creates a universal digital twin platform that can represent multiple types of mobility entities including vehicles, pedestrians, cyclists, and traffic infrastructure through a common digital twin architecture. The platform provides multi-functional capabilities: data ingestion from diverse sources, standardized data schema validation, machine learning for predictive analytics, simulation for scenario testing, and actuation for control. This universal approach integrates human, vehicular, and traffic entities within a single manageable framework rather than requiring separate systems for each entity type.
Data Source
AI summary
Systems and methods are provided for effectuating and using a mobility digital twin (MDT) framework including a digital space (where digital twins reside) and a physical space (where physical objects/processes reside). The MDT framework is realized in a cloud-based system/on a cloud platform. Digital twins may represent not only vehicular entities, but human and traffic entities as data/models representative of these different physical objects/processes may be applicable to more than just a directly-related entity. Additionally, the MDT framework is able to leverage data associated with different time horizons (e.g., real-time data as well as historical data).