Autonomous Vehicle Ride Quality Assessment Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in identifying and mitigating events that cause passenger discomfort, and assessing overall ride quality, which affects user experience and retention.
Innovation Solution
Training models using ride data from both actual and simulated rides, incorporating feedback from passengers to identify uncomfortable events and overall ride quality, allowing for software improvements and intervention strategies.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If ride quality assessment relies on manual passenger feedback and review, then measurement precision is limited and time-consuming, but automation extent is low
Solution Approach 1:
The system enables automated self-assessment of ride quality by training machine learning models to independently evaluate ride data, identify uncomfortable events, and generate assessments without requiring manual passenger review for each ride
Solution Approach 2:
The system implements feedback loops where passenger ratings and feedback are continuously collected, used to train and improve the machine learning models, which then generate automated assessments that can be compared against actual passenger experience to refine future predictions
2Measurement precision
If comprehensive ride data collection is implemented to improve measurement precision, then loss of information is reduced, but device complexity increases
Solution Approach 1:
The machine learning models serve multiple functions: they identify uncomfortable events, assess overall ride quality, prioritize engineering work, and generate actionable insights, allowing a single data collection system to support diverse assessment needs without requiring separate specialized systems for each function
Solution Approach 2:
The assessment system is divided into modular components including event identification models, overall quality assessment models, and prioritization algorithms, allowing the complex data collection and analysis task to be broken into manageable segments that can be developed and maintained independently
3Productivity
If automated model training is used to improve productivity, then loss of time is reduced, but measurement precision may be compromised without sufficient training data
Solution Approach 1:
The system performs preliminary actions by continuously training models on accumulated ride data in the background, so that when new ride data needs assessment, the models are already prepared and can process quickly without compromising accuracy through adequate pre-training
Solution Approach 2:
The model training process operates continuously rather than in discrete batches, allowing the system to constantly learn from new data and maintain high measurement precision while processing assessments at high speed through uninterrupted learning and adaptation
Data Source
AI summary
The disclosure relates to training models for identification of events likely to cause discomfort to passengers of autonomous vehicles and for assessment of overall ride quality of autonomous vehicle rides. For instance, ride data may be associated with a ride quality value indicative of a level of discomfort and/or a first overall ride quality value indicating an overall ride quality provided by the passenger for the first ride. This ride data may be used to train a model such that the model is configured to, in response to receiving ride data for a second ride as input, output a list of events likely to cause discomfort to a passenger during the second ride and/or such that the model is configured to, in response to receiving second ride data for a second ride as input, output a second overall ride quality value for the second ride.