Vehicular Radar Object Classification via Reflection Response Matching

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current vehicle sensing systems using radar sensors are limited in identifying vehicle attributes such as type or model, relying solely on range and velocity information, which hinders decision-making in safety applications like autonomous driving.

Innovation Solution

The system employs high-definition radar sensors with multiple transmitters and receivers on an antenna array to capture detailed Radar Reflection Responses, which are compared to stored data sets to classify detected vehicles by type or model, integrating this information with machine vision data for enhanced decision-making.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional radar sensors are used to provide basic range and velocity information, then the system complexity is low, but the measurement precision of vehicle attributes (type, model) is insufficient

Engineering Contradiction:
Improvevehicle attribute identification accuracyVSAvoidsensor system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The radar antenna array is segmented into multiple independent transmitters and receivers, allowing each element to contribute to forming distinct radar reflection response signatures. This segmentation enables fine-grained spatial resolution and detailed vehicle attribute identification without requiring a completely new sensor system.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system transitions from conventional 1D range-velocity measurement to 2D angular resolution (azimuth and elevation) by utilizing the spatial distribution of multiple antenna elements. This dimensional expansion enables the system to capture vehicle attributes such as type and model through geometric relationships in the radar reflection patterns.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

2Measurement precision

If high-definition radar sensors with multiple transmitters and receivers are used to capture detailed Radar Reflection Responses, then the measurement precision of vehicle attributes improves, but the device complexity increases

Engineering Contradiction:
ImproveRadar Reflection Response resolutionVSAvoidantenna array complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The high-definition radar sensor with multiple transmitters and receivers serves multiple functions: it provides conventional range and velocity measurement, captures detailed radar reflection response patterns, and enables vehicle attribute classification. This multi-functionality justifies the increased device complexity by eliminating the need for separate sensing systems.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system changes the operational parameters of the radar sensor by utilizing multiple transmitters and receivers simultaneously, transforming the output from simple range-velocity data to detailed radar reflection response datasets. This parameter change enables high-resolution vehicle attribute measurement while maintaining a single sensor platform.

Inventive Principle:
Principle #35Parameter changes

3Reliability

If radar reflection response data is collected and compared to stored data sets for vehicle classification, then the reliability of safety applications improves, but the processing time and computational complexity increase

Engineering Contradiction:
Improvesafety application reliabilityVSAvoiddata processing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

Radar reflection response data for various vehicle types and models is pre-collected and stored in databases before actual operation. During runtime, the system only needs to compare captured radar patterns against these pre-stored references, significantly reducing real-time computational requirements while maintaining high reliability in vehicle classification for safety applications.

Inventive Principle:
Principle #10Preliminary action

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

This approach enables early identification of vehicle attributes, improving the reliability of safety applications by providing high-definition location and velocity data, allowing for accurate classification and enhanced decision-making in autonomous vehicle operations.

Implementation Method 1

The at least one radar sensor comprises multiple Tx (transmitters) and Rx (receivers) on an antenna array, so as to provide high definition, fine resolution in azimuth and/or elevation to determine high definition Radar Reflection Responses for objects detected by the system.

Methodology Applied
Scientific EffectRadar: Radar

Implementation Method 2

The system generates a data set of radar reflection responses for an object in the field of sensing of said at least one radar sensor

Methodology Applied
Scientific EffectReflection: Reflection

Data Source

PatentUS11604253B2Vehicular sensing system for classification of detected objects
Publication Date: 2023.03.14 MAGNA ELECTRONICS INC
  • US11604253B2 patent drawing
  • US11604253B2 patent drawing
  • US11604253B2 patent drawing

AI summary

A vehicular sensing system includes at least one radar sensor disposed at a vehicle and having a field of sensing forward, rearward or sideward of the vehicle. Radar data captured by the radar sensor is received at an electronic control unit (ECU). Received transmitted signals reflected off objects and received at the receiving antennas are evaluated at the ECU to establish surface responses for the objects present in the field of sensing of the radar sensor. A data set of radar data that is representative of an object present in the field of sensing of the radar sensor is compared to stored data sets to determine if the data set corresponds to a particular stored data set of the stored data sets. Responsive to the data set of radar data being determined to correspond to the particular stored data set, the vehicular sensing system classifies the detected object.