Adaptive Vehicle Data Sampling for ML Driving Issue Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle data collection techniques face challenges in balancing data quantity, sensor capabilities, and network constraints, leading to reduced data collection and wastage of computing and networking resources due to limited bandwidth and battery life of sensors.
Innovation Solution
A vehicle platform that dynamically adjusts sampling rates and time periods based on network and device constraints to collect and process vehicle data, using machine learning models to identify driving behaviors or issues, thereby optimizing data collection and resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If data collection is greatly reduced to conserve resources, then resource wastage is reduced, but measurement precision and detection capability deteriorate
Solution Approach 1:
The system dynamically changes sampling parameters (sampling rates, time periods, data collection frequency) based on learned patterns from historical data. By adjusting these parameters adaptively, the system collects sufficient data to maintain detection precision while minimizing resource consumption during normal operation.
Solution Approach 2:
The machine learning model enables the system to automatically determine optimal data collection strategies without external intervention. The model learns from historical data what conditions require detailed monitoring versus when reduced monitoring is sufficient, allowing the system to self-optimize its resource usage while maintaining detection capability.
2Productivity
If sampling rates are reduced to conserve network and computing resources, then resource utilization improves, but information completeness deteriorates
Solution Approach 1:
The system performs preliminary analysis of historical vehicle data to establish baseline patterns and anomaly thresholds before actual monitoring begins. This preliminary learning phase enables the system to distinguish between normal variations and significant events, allowing reduced sampling during normal conditions while maintaining high sampling rates when anomalies are detected.
Solution Approach 2:
The machine learning model continuously receives feedback from incoming data streams and adjusts sampling rates dynamically. When the model detects patterns indicating potential issues or when data deviation from learned norms exceeds thresholds, it automatically increases sampling frequency to capture complete information, then returns to lower sampling rates when conditions normalize.
3Duration of action of moving object
If data collection is reduced to extend battery life of sensors, then device durability improves, but detection reliability deteriorates
Solution Approach 1:
The system implements periodic data collection at variable intervals rather than continuous sampling. The machine learning model determines optimal sampling intervals based on learned patterns, creating a rhythm of high and low activity that extends battery life while ensuring sufficient data is collected to maintain detection reliability through periodic full-scale monitoring.
Data Source
AI summary
A device receives network constraints associated with a network connected to a vehicle device, data collection constraints, and vehicle device constraints. The device determines a first sampling rate, a first time period, a second sampling rate, and a second time period for collecting vehicle data based on the network constraints, the data collection constraints, and the vehicle device constraints, wherein the first sampling rate is different than the second sampling rate. The device receives first vehicle data, provided at the first sampling rate and for the first time period, and second vehicle data, provided at the second sampling rate and for the second time period, and processes the first and second vehicle data, with a machine learning model, to identify a driving behavior or a vehicle issue. The device performs one or more actions based on the driving behavior or the vehicle issue.