Vehicle Cabin Event Detection Using Gas and PM Time-Series
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In shared vehicle services, enforcing rules and policies regarding vehicle usage is challenging due to the lack of direct interaction between users and operators, leading to difficulties in detecting and addressing violations autonomously.
Innovation Solution
A system comprising gas sensors, particulate matter sensors, and a controller using machine learning models to detect and classify events in a vehicle cabin, such as smoking, vaping, or cleanliness issues, by analyzing sensor data and transmitting alerts to operators, reducing the need for human intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If autonomous detection systems are implemented in shared vehicles, then the need for human intervention is reduced, but the complexity of the detection system increases
Solution Approach 1:
The patent combines multiple sensor types (gas sensors, particulate matter sensors, temperature sensors, humidity sensors) into a unified monitoring system that collectively detects various vehicle conditions. This merging of sensing capabilities enables comprehensive autonomous detection while managing system complexity through integrated architecture.
Solution Approach 2:
The monitoring system is designed to detect multiple types of events simultaneously (smoking, vaping, spills, temperature extremes, humidity issues) using a single integrated platform. This multi-functional approach reduces the need for separate specialized systems, thereby managing complexity while enhancing automation.
2Measurement precision
If multiple sensors are deployed to detect various events, then detection accuracy improves, but system complexity and cost increase
Solution Approach 1:
The system divides detection responsibilities across multiple specialized sensors, each optimized for specific parameters (gas composition, particulate matter, temperature, humidity). This segmentation allows high detection accuracy for each event type while managing overall complexity through modular sensor design and dedicated processing for each sensor type.
Solution Approach 2:
The controller serves as an intermediary that receives data from multiple sensors, processes the information, and determines events based on combined inputs. This intermediary processing layer integrates signals from various sensors, enabling accurate multi-parameter detection while simplifying the complexity through centralized coordination.
3Reliability
If continuous monitoring is implemented, then event detection reliability improves, but energy consumption increases
Solution Approach 1:
The system implements periodic sampling of sensor data rather than truly continuous monitoring, with the controller periodically reading sensor values and analyzing changes. This periodic approach maintains detection reliability by capturing event occurrences while reducing energy consumption compared to constant high-rate sampling.
Solution Approach 2:
The system includes self-diagnostic capabilities where the controller monitors sensor functionality and system status, automatically managing its own operation. This self-service approach optimizes energy usage by adjusting monitoring intensity based on detected conditions and system state, maintaining reliability while minimizing unnecessary energy consumption.
Data Source
AI summary
A system for determining an event in a cabin of a vehicle includes a gas sensor configured to generate a first sensor signal associated with a quantity of at least one gas or VOC in ambient air of the cabin, a PM sensor configured to generate a second sensor signal associated with a quantity of particulate matter in the ambient air of the cabin, and a controller operably connected to the gas sensor and the PM sensor. The controller is configured to receive the first and second sensor signals, generate first and second time series datasets of the first and second sensor signals, and determine an event in the cabin of the vehicle by analyzing both the first and second time series datasets using a machine learning model that has been trained with training data corresponding to time series data of PM readings and gas sensor readings of known events.