Vehicle Object Detection Clustering Spatial Velocity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle object detection systems using Doppler sensors, such as radar and lidar, face inefficiencies in identifying objects quickly due to reliance on temporal techniques like the Kalman filter, which require data over a specified period before object identification can occur.
Innovation Solution
Implementing a clustering algorithm that considers both spatial proximity and velocity measurements from multiple Doppler sensors to generate clusters of detection points, allowing for immediate object identification without waiting for a temporal cycle, thereby enhancing detection speed and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If temporal techniques like Kalman filter are used for object detection, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent segments the object detection process into spatial clustering and velocity filtering stages. Detection points are first grouped into clusters based on spatial proximity, then velocity information is used to filter and identify objects within those clusters. This segmentation allows parallel processing of spatial and temporal data, reducing overall detection time while maintaining accuracy.
Solution Approach 2:
The patent transitions from traditional temporal-only filtering (Kalman filter) to a two-dimensional approach combining spatial clustering with velocity-based temporal filtering. By adding the spatial dimension as a preliminary grouping step, the system can process multiple detection points simultaneously and apply velocity filtering more efficiently, reducing the time delay inherent in sequential temporal processing.
2Measurement precision
If multiple sensors are used to collect detection data, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent merges detection points from multiple sensors into unified spatial clusters before applying velocity-based object identification. By combining spatial information from all sensors into clusters first, then applying a unified velocity filtering criterion, the system processes multi-sensor data more efficiently than separate processing streams, reducing computational complexity while maintaining the precision benefits of multiple sensors.
Solution Approach 2:
The patent creates a universal clustering framework that handles detection points from multiple sensor types and configurations through a common spatial-grouping mechanism. This universal approach allows the same clustering and velocity-filtering logic to process data from various sensor arrangements, reducing the need for sensor-specific processing code and simplifying the overall system architecture.
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 faster identification of objects by clustering detection points based on spatial and velocity thresholds, allowing the vehicle's computer to recognize and avoid objects more quickly than traditional temporal methods, such as the Kalman filter, by processing data in real-time.
Implementation Method 1
sensors to collect data about a surrounding environment... each data set including a range, an azimuth angle, and a range rate for a detection point... determine, for each detection point, a radial component of a ground speed of the detection point based on the data set associated with the detection point
Data Source
AI summary
A computer includes a processor and a memory storing instructions executable by the processor to collect a plurality of data sets, each data set from a respective sensor in a plurality of sensors, and each data set including a range, an azimuth angle, and a range rate for a detection point of the respective one of the sensors on an object to determine, for each detection point, a radial component of a ground speed of the detection point based on the data set associated with the detection point and a speed of a vehicle, and to generate a plurality of clusters, each cluster including selected detection points within a distance threshold from each other and having respective radial components of ground speeds that are (1) above a first threshold and (2) within a second threshold of each other.