Vehicular Micro Clouds for Low-Latency Anomaly Mapping

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current vehicles rely on onboard sensors for anomaly detection, resulting in incomplete anomaly maps due to limited perspectives, and traditional Vehicle-to-Everything (V2X) communication faces issues like latency and underdeveloped infrastructure, making it unsuitable for effective anomaly mapping.

Innovation Solution

The implementation of a vehicular micro cloud system, where an anomaly client and detector cooperate to generate improved anomaly maps by forming stationary or mobile vehicular micro clouds, allowing vehicles to share anomaly data and coordinate sensors and resources to enhance anomaly detection and prediction.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If traditional V2X communication is used to improve anomaly map completeness, then multiple vehicle perspectives can be integrated, but latency and communication overhead increase

Engineering Contradiction:
Improveanomaly map completenessVSAvoidcommunication latency
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The system segments the anomaly mapping task by having each vehicle independently detect anomalies and create local anomaly maps using its onboard sensors. Each vehicle maintains its own anomaly detection capabilities rather than relying on centralized data collection, thus reducing communication latency while preserving information completeness through distributed mapping.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Vehicles create local copies of anomaly maps based on their own sensor data and perspectives. Instead of one vehicle requesting data from others through V2X communication, each vehicle independently generates and maintains its own anomaly map copy, eliminating the need for continuous data exchange while ensuring each has complete local information.

Inventive Principle:
Principle #26Copying

2Loss of information

If traditional V2X communication is used to share anomaly data between vehicles, then comprehensive anomaly maps can be generated, but infrastructure requirements increase

Engineering Contradiction:
Improveanomaly detection accuracyVSAvoidinfrastructure complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

Each vehicle performs self-service anomaly detection using its own onboard sensors and processing capabilities. Vehicles independently detect anomalies, generate anomaly maps, and maintain their own anomaly databases without requiring external infrastructure support. This eliminates the need for complex V2X communication infrastructure while maintaining high anomaly detection accuracy through autonomous operation.

Inventive Principle:
Principle #25Self-service

3Device complexity

If single vehicle onboard sensors are used for anomaly detection, then system complexity is minimized, but anomaly map quality deteriorates

Engineering Contradiction:
Improvesystem simplicityVSAvoidanomaly map quality
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The system segments the anomaly detection function across multiple vehicles, where each vehicle maintains simple onboard sensors but collectively provides comprehensive coverage. Each vehicle's simple sensor system contributes to a distributed network that achieves high-quality anomaly mapping through multiple perspectives without requiring complex individual vehicle systems.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Multiple vehicles merge their individual anomaly map data to create comprehensive anomaly maps of the environment. By combining the simple anomaly detection capabilities of multiple vehicles, the system achieves high measurement precision and complete environmental coverage while each vehicle maintains device simplicity.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentEP3694237B1Anomaly mapping by vehicular micro clouds
Publication Date: 2023.09.13 TOYOTA JIDOSHA KK
  • EP3694237B1 patent drawingFigure 1A
  • EP3694237B1 patent drawingFigure 1B
  • EP3694237B1 patent drawingFigure 1C

AI summary

The disclosure includes embodiments for generating improved anomaly maps. In some embodiments, a method for a connected vehicle includes detecting an occurrence of an anomaly in a roadway environment based on sensor data describing the roadway environment. The method includes creating, by the connected vehicle, an anomaly map that describes the anomaly. The method includes modifying an operation of a communication unit of the connected vehicle to receive one or more other anomaly maps describing the anomaly from one or more cooperation endpoints in the roadway environment. The method includes generating an updated anomaly map based on the anomaly map created by the connected vehicle and the one or more other anomaly maps created by the one or more cooperation endpoints so that an accuracy of the updated anomaly map is improved.