FMCW Radar Occupancy Detection Using Feature Vector Classification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing radar systems for vehicle occupancy detection face challenges in accurately distinguishing occupants from other moving objects, are prone to delayed responses, and struggle in low ambient light conditions, especially when objects are covered or in complex environments.
Innovation Solution
A frequency modulated continuous wave (FMCW) radar sensing system using feature vectors, which includes sensors, an analogue-to-digital converter, and a processor to generate point cloud datasets, extract features, and classify objects based on range, angle, velocity, and reflected power, enabling faster and more accurate detection and differentiation of occupants in various vehicle zones.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If radar systems use vital signs for occupancy detection, then detection accuracy is improved, but response time is delayed
Solution Approach 1:
The system performs preliminary classification of detected objects using multiple features (range, angle, velocity, reflected power, point cloud characteristics) before final occupancy determination. This preliminary sorting and filtering of objects based on their radar characteristics allows the system to quickly identify potential occupants without waiting for vital sign analysis, thus reducing response time while maintaining accuracy through subsequent verification.
2Measurement precision
If radar sensors detect smallest motions, then occupancy detection capability is improved, but noise discrimination becomes difficult
Solution Approach 1:
The system segments the radar detection process into multiple independent analysis stages: signal acquisition, point cloud generation, feature extraction (range, angle, velocity, reflected power), object classification, and noise filtering. By dividing the detection process into discrete segments with specific functions, the system can apply targeted noise discrimination techniques at each stage rather than attempting to filter noise from raw signals, thereby improving both sensitivity and noise rejection.
Solution Approach 2:
The system introduces point cloud data and multiple intermediate features as mediators between the raw radar signal and the final occupancy detection. These intermediate representations serve as a buffer that allows the system to analyze object characteristics from multiple perspectives (spatial, temporal, spectral) before making a detection decision, effectively separating the signal of interest from background noise and interference.
3Device complexity
If one sensor covers multiple seats, then system complexity is reduced, but detection precision deteriorates
Solution Approach 1:
The system transitions from spatial segmentation (multiple sensors for multiple seats) to feature-space segmentation (multiple features from single sensor). By extracting and analyzing multiple dimensions of information from each detected object (range, angle, velocity, reflected power, point cloud density, temporal patterns), the system can differentiate between objects in different seats using a single sensor, achieving seat-specific detection without requiring multiple physical sensors.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The system ensures rapid occupancy detection, can operate under low ambient light and with covered objects, and effectively distinguishes between occupants and other objects, providing robust in-cabin features like life presence detection and adult/child classification.
Implementation Method 1
frequency modulated continuous wave (FMCW) radar sensing for classification of objects in a vehicle
Implementation Method 2
a transmitter unit that emits signals onto the object, and a receiver unit that receives the emitted signals reflected from the object
Data Source
AI summary
The present disclosure relates to a system for differentiating objects present in a vehicle, the system includes one or more sensors placed within a vehicle to generate a set of signals in response to an object being present within the vehicle. An ADC converts the received set of signals to a digital form. A processor receives the digital set of signals, and process the received digital set of signals, to generate point cloud dataset. The processor extracts, from the point cloud dataset, a first set of features pertaining to a single frame and a second set of features pertaining to a multi-frame. The extracted set of features are provided as input to a classifier to differentiate the object present in one or more zones within the vehicle.