Distributed Vehicle Sensor Data Analysis for Latency and Thermal Constraints
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In the automotive field, real-world vehicle event data is sparse and difficult to isolate due to redundancy and irrelevant information in datasets, and there are challenges in processing and storing data at the edge with limited computing power, leading to issues with accuracy and latency in both edge and cloud computing.
Innovation Solution
A distributed data analysis method that collects and processes vehicle sensor data using both edge and remote computing subsystems, reducing redundancy and computational demands at the edge while enhancing accuracy and reducing latency by dividing processing responsibilities between the two systems.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data is processed and stored in the cloud, then computational accuracy is improved, but communication latency and bandwidth costs increase
Solution Approach 1:
The patent segments the data processing system into edge computing devices deployed in the field and cloud computing resources. Edge devices perform initial data collection, filtering, and preprocessing locally, while the cloud handles more complex analysis. This segmentation allows critical processing to occur close to data sources (reducing latency) while maintaining cloud-based computational accuracy for non-time-critical operations.
Solution Approach 2:
The patent implements preliminary action by performing data filtering, aggregation, and preprocessing at the edge computing devices before transmitting data to the cloud. This preliminary processing reduces the volume and complexity of data requiring cloud transmission, thereby reducing communication latency and bandwidth requirements while preserving the accuracy benefits of cloud-based analysis.
2Speed
If computing power is increased at the edge, then processing speed is improved, but thermal and power constraints are exceeded
Solution Approach 1:
The patent segments computational tasks between edge computing devices and cloud systems based on their respective capabilities. Edge devices perform lightweight, time-critical processing that respects their thermal and power constraints, while more computationally intensive tasks are offloaded to the cloud. This segmentation enables processing speed improvement for critical functions without exceeding edge device thermal limits.
Solution Approach 2:
The patent applies partial action by implementing a tiered processing approach where only essential, time-critical computations are performed at the edge with limited power, while less time-sensitive but more computationally demanding tasks are performed partially or fully in the cloud. This allows the system to achieve necessary processing speeds for critical operations without over-provisioning edge computing power that would violate thermal constraints.
3Quantity of substance
If all real-world data is collected and analyzed, then data completeness is improved, but data redundancy and irrelevant information increase
Solution Approach 1:
The patent extracts and removes redundant and irrelevant information through intelligent filtering mechanisms implemented at edge computing devices. The system identifies and extracts only the most relevant features and data points from raw sensor inputs, discarding duplicates and irrelevant information before transmission to the cloud. This extraction process maintains data completeness for critical information while eliminating redundancy.
Solution Approach 2:
The patent performs preliminary data filtering, aggregation, and relevance assessment at the edge computing devices before cloud transmission. This preliminary action identifies and retains only essential data elements, removing redundancy and irrelevant information in advance. The result is a streamlined dataset that maintains completeness for analysis-critical information while significantly reducing overall data volume and redundancy.
Data Source
AI summary
Systems and methods for distributed event detection. Sensor data is synchronized, and an on-board detector included in a vehicle detects an event from the synchronized sensor data. The synchronized sensor data is transmitted to a remote system. At the remote system, a remote detector detects the event and generates a remote label from at least a data subset of the synchronized sensor data.