Vehicle Software Update Actuation Using Causal Telemetry Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Software updates for vehicles can have varying effects due to complex interactions and operating conditions, potentially leading to adverse impacts on vehicle performance, especially in electric vehicles, and may be compounded by simultaneous updates.
Innovation Solution
A system using causal machine learning to analyze vehicle data and actuate changes, such as reverting software updates or disabling components, based on the effectiveness of the updates to maintain vehicle performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If software updates are pushed to vehicles to address performance issues, then vehicle performance is improved, but adverse effects may occur due to complex interactions and varying operating conditions
Solution Approach 1:
The system performs preliminary monitoring and evaluation of software update effectiveness before fully deploying updates across the fleet. By continuously collecting telemetry data and assessing performance metrics after updates are applied, the system can identify adverse effects early and prevent widespread deployment of problematic updates, thus resolving the contradiction between improving performance and avoiding harmful effects.
Solution Approach 2:
The system implements a feedback mechanism that continuously monitors vehicle performance metrics after software updates are applied. Telemetry data is collected and analyzed to determine whether updates achieve their intended performance improvements or cause adverse effects. This feedback loop enables dynamic decision-making about update deployment, allowing the system to maintain reliability while minimizing harmful effects by reverting or withholding updates that cause problems.
2Productivity
If multiple software updates are pushed simultaneously to vehicles, then comprehensive performance improvements are achieved, but the complexity of interactions increases and may cause adverse effects
Solution Approach 1:
The system segments the software update deployment process by evaluating and monitoring each update's individual effectiveness through telemetry data analysis. Instead of treating multiple simultaneous updates as a single complex intervention, the system breaks down the assessment into individual update evaluations, tracking their separate contributions to performance changes. This segmentation reduces the complexity of understanding interactions while still achieving comprehensive performance improvements.
3Speed
If software updates are deployed without thorough testing across all operating conditions, then deployment speed is increased, but effectiveness varies due to complex interactions and varying conditions
Solution Approach 1:
The system enables self-service evaluation of software update effectiveness by automatically collecting telemetry data from vehicles and analyzing performance metrics without requiring manual testing across all operating conditions. The system serves itself by using real-world operational data to assess whether updates achieve their intended effects, maintaining rapid deployment speed while improving measurement precision through continuous post-deployment monitoring and analysis.
Data Source
AI summary
A computer that includes a processor and a memory, the memory including instructions executable by the processor to receive data corresponding to a plurality of vehicles regarding a specified aspect of vehicle performance. An effectiveness of a software update targeted to the specified aspect of vehicle performance is determined based on the data, and when the determined effectiveness is below a specified threshold the system actuates a change in at least one of the plurality of a vehicles.