Detecting Automatic Driving via Heuristic Vehicle Data Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional automobile insurance premium pricing fails to consider the risk reduction from advanced driver assistance systems (ADAS) like autopilot and self-driving technologies, as direct data on their engagement is often not available or accessible, leading to inaccurate premium calculations.
Innovation Solution
A system that detects the engagement of automatic driving features by analyzing sequential vehicle operation data using heuristic rules, generating automatic driving information without explicit indication, allowing for individualized and real-time insurance premium adjustments based on actual usage of ADAS capabilities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct data on ADAS engagement is obtained from vehicle systems, then measurement precision of automatic driving usage is improved, but device complexity and data accessibility worsen due to restricted access to vehicle computer systems
Solution Approach 1:
The patent introduces an intermediary detection system that observes vehicle operation data (speed, power consumption, heading, location) as a mediator to infer ADAS engagement status without directly accessing the vehicle's internal computer systems. This intermediary approach bridges the gap between the need for accurate ADAS detection and the inability to directly access vehicle systems.
Solution Approach 2:
The patent replaces direct electronic access to vehicle systems with an observational analysis method that uses heuristic rules on operational parameters. Instead of mechanically or electronically connecting to the vehicle's computer system, the system substitutes this with analytical processing of publicly available operational data to detect ADAS engagement.
2Ease of operation
If heuristic rules are used to detect ADAS engagement from vehicle operation data, then ease of operation and data accessibility are improved, but measurement precision of automatic driving detection deteriorates
Solution Approach 1:
The patent applies partial action by using a subset of vehicle operation parameters (speed, power consumption, heading, location) rather than requiring complete access to all vehicle systems. The heuristic rules analyze only the necessary operational indicators to detect ADAS engagement, performing sufficient detection without excessive data collection or processing complexity.
Solution Approach 2:
The patent transforms the detection approach by changing from direct system state parameters to operational performance parameters. Instead of reading direct ADAS engagement flags from vehicle systems, the system monitors changes in operational parameters (speed patterns, power consumption levels, heading stability, location tracking) that correlate with ADAS activation, using these parameter changes as indirect indicators.
3Ease of manufacture
If traditional insurance pricing based on aggregated statistical data is used, then ease of manufacture and pricing simplicity are improved, but reliability and accuracy of risk assessment worsen due to lack of individualized usage data
Solution Approach 1:
The patent introduces dynamic pricing capability by enabling real-time or near-real-time detection of ADAS engagement during vehicle operation. Instead of static, annual pricing based on historical aggregates, the system dynamically adjusts pricing based on actual, observed usage patterns, making the insurance system adaptive to individual driver behavior and vehicle technology utilization.
Solution Approach 2:
The patent implements a feedback loop where vehicle operation data is continuously collected, analyzed through heuristic rules to detect ADAS engagement, and then used to adjust insurance pricing. This feedback mechanism connects actual usage behavior back to pricing decisions, enabling more accurate risk assessment that reflects individualized usage patterns rather than population averages.
Data Source
AI summary
Activation of an automatic driving feature in a vehicle is detected by evaluating sequential vehicle operation data against subtractive and additive heuristic rules that define a likelihood of an automatic driving feature having been engaged as a function of vehicle performance. The sequential vehicle operation data, which does not include an explicit indication of whether the automatic driving feature was engaged, is provided for time intervals of a trip made by the vehicle. Automatic driving information is generated, which provides an indication of whether the automatic driving feature was engaged during a subset of the time intervals.