Radar Object Orientation via Spatial Distribution Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current radar-based object detection systems cannot accurately determine the type of object and its spatial orientation, which limits the prediction of future trajectories and collision avoidance in automated driving, especially under low-light conditions.
Innovation Solution
A method to determine the spatial orientation of an object from measuring signals responding to electromagnetic interrogation radiation, using a classifier or regressor to analyze the spatial distribution of contributions from different locations on the object, allowing for improved prediction of object trajectories and potential collision risks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If radar sensors aggregate raw data into reflections and extract only angular position, intensity, and distance, then the processing complexity is reduced, but the spatial extension and orientation information of detected objects is lost
Solution Approach 1:
The patent extracts the spatial distribution of contributions from different locations on the object by inverting the radar signal processing approach. Instead of aggregating all raw data into single reflections, the method identifies and separates contributions from different spatial locations, creating a detailed spatial map that preserves orientation information while still using standard radar processing for the overall signal.
2Reliability
If conventional radar systems classify objects using machine learning based on aggregated reflection data, then false alarms are reduced, but the ability to determine spatial orientation and object type is insufficient
Solution Approach 1:
The patent segments the object's surface into multiple contribution locations, each contributing to the overall radar reflection. By analyzing the spatial distribution of these segmented contributions rather than treating the object as a single reflected entity, the system achieves precise spatial orientation measurement while maintaining reliable object classification through machine learning on the enhanced spatial data.
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
Enhances the accuracy of predicting object trajectories and reduces collision risks by determining the spatial orientation of objects using machine learning classifiers or regressors, even in low-light conditions, thereby improving the safety of automated driving systems.
Implementation Method 1
the response of the object to electromagnetic interrogation radiation... This response may include a reflection of the interrogation radiation
Data Source
AI summary
A method for determining the spatial orientation of an object from at least one measuring signal which includes the response of the object to electromagnetic interrogation radiation. A method for predicting the trajectory of at least one object from at least one measuring signal which includes the response of the object to electromagnetic interrogation radiation, in conjunction with a scalar velocity of the object. A method for training a classifier and/or a regressor.