Autonomous Vehicle State Change Assessment After Remote Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous devices, such as vehicles and IoT systems, face challenges in accurately assessing the impact of software and hardware changes on their behavior, leading to difficulties in maintenance scheduling, liability insurance, and optimal operation due to complex AI control code bases and frequent updates.
Innovation Solution
A method and system that evaluates performance parameter values from autonomous devices, incorporating information about software and hardware changes to compute a performance quantity, allowing for the determination of a change of state value that indicates the effect of these changes on the device's operation, enabling adaptive maintenance and insurance management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If frequent remote updates are applied to autonomous devices, then device functionality and adaptability are improved, but reliability assessment and evaluation become difficult
Solution Approach 1:
The patent implements a feedback mechanism where performance data from autonomous devices is continuously collected and used to update performance models. This feedback loop enables reliable assessment of device state changes even after frequent updates by comparing actual performance against updated expectations.
Solution Approach 2:
The patent changes the parameter of assessment from static baseline comparisons to dynamic performance modeling that adapts to software and hardware changes. By updating performance models to reflect current device configurations, reliable assessment becomes possible despite frequent updates.
2Measurement precision
If conventional statistical testing is used for evaluation, then systematic assessment is achieved, but the frequency of patches/updates interferes with testing validity
Solution Approach 1:
The patent transitions from static statistical testing to a dynamic performance modeling approach. The performance models are continuously updated to reflect current device states, allowing evaluation to remain valid despite frequent updates. This dynamic adaptation eliminates the time loss associated with invalidating test baselines.
3Adaptability or versatility
If complex AI control code bases are used, then autonomous device capabilities are enhanced, but quality management burden increases
Solution Approach 1:
The patent extracts the complexity management problem from the AI code base itself and handles it separately through performance modeling. By measuring external performance parameters rather than analyzing internal code complexity, the system manages quality without being burdened by code base complexity.
4Productivity
If remote updates are applied to IoT devices, then device optimization is achieved, but evaluation reliability becomes difficult
Solution Approach 1:
The patent performs preliminary actions by establishing performance models before updates and then updating these models after changes. This preliminary and follow-up modeling ensures that evaluation reliability is maintained throughout the optimization process, allowing continuous improvement without losing assessment accuracy.
Data Source
Figure 1~2
Figure 3~4
Figure 5
AI summary
The present invention relates to a method (100) and a system (1) for determining a change of state of an autonomous device (2), in particular an autonomous vehicle. Therein, a plurality of performance parameter values (v) obtained by monitoring at least one performance parameter during autonomous operation of the device (2) is received. Further, a performance quantity (q) quantifying the quality of autonomous operation of the device (2), in particular the quality of driving of the autonomous vehicle, based on the obtained performance parameter values (v) and information (i) associated with a flux of software and/or hardware related to autonomous operation of the device (2) is determined. Further, a change of state value for the device (2) is determined based on the performance quantity (q).