Driving Intention Prediction for Cold Start Emissions Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Evolving emissions regulations for light duty vehicles require aftertreatment devices to be activated quickly, especially during cold starts when engine heat is insufficient, necessitating auxiliary heating devices that need to be activated before vehicle operation to control emissions effectively.
Innovation Solution
A system and method for predicting driving intention by monitoring sequences of indicators such as door openings, key status, and seatbelt engagement, determining separation times, and comparing them to historical data to calculate a confidence level for engine start, thereby activating cold start and aftertreatment devices proactively.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-generated harmful factors
If auxiliary heating devices are activated before vehicle start, then emissions control is improved during cold start, but battery energy is consumed before engine operation
Solution Approach 1:
The system performs preliminary action by detecting sequences of driver intentions (door unlock, door open, seatbelt engagement) and activating auxiliary heating devices before the engine actually starts. This preliminary activation ensures emissions control is ready when the engine starts cold, while the intelligent sequencing based on detected driver actions helps optimize when energy is consumed.
Solution Approach 2:
The system uses feedback by continuously monitoring multiple indicators of driver intention (door locks, door switches, seatbelt buckles) and adjusting the activation timing of heating devices based on the detected sequence. The historical data comparison provides feedback on typical driver behaviors, allowing the system to predict when engine start is likely and activate heating at the optimal moment to balance emissions control with battery conservation.
2Loss of time
If multiple indicators are monitored to predict driving intention, then activation timing is improved, but system complexity increases
Solution Approach 1:
The system segments the driving intention detection into multiple discrete indicator events (door unlock, door open, seatbelt engage) that can be independently detected and sequenced. Each indicator is a separate detectable event, and the system processes them in sequence to build confidence in the prediction, breaking down the complex task of intention recognition into manageable segments.
Solution Approach 2:
The system uses universal vehicle components and sensors that already exist in modern vehicles (door locks, door switches, seatbelt buckles) for the additional function of intention prediction. These existing components serve their primary safety and operational functions while also providing data for the prediction algorithm, avoiding the need for dedicated new sensors or complex additional hardware.
Data Source
AI summary
Disclosed are methods, systems, and computer-readable mediums for predicting a driving intention of a driver of a vehicle. A sequence comprising a plurality of indicators is detected, where each indicator suggests intent to drive a vehicle. A separation time between each of the indicators of the sequence is determined. The sequence and determined separation times are compared to historical data, where the historical data comprises data related to previously stored separation times of the sequence. Based on the comparison, a confidence level that an engine of the vehicle will be started is determined. Based on the confidence level, a feature of the vehicle is activated.