Radar Object Tracking with Bounding-Shape Kalman Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object tracking methods using Kalman filters face inaccuracies due to large uncertainties in velocity vector estimates and inclusion of noisy sensor measurements, particularly when tracking objects with multiple portions, leading to reduced accuracy and increased computational costs.
Innovation Solution
The method incorporates angle measurements alongside range rate measurements to determine expected range rates and positions based on a bounding shape, using a Kalman filter to improve tracking accuracy by accounting for differences in range rates and noise levels across object portions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional Kalman filter updates are performed sequentially with multiple sensor measurements corresponding to different portions of an object, then all measurements are processed to determine state estimates, but the velocity vector estimates have large levels of uncertainty due to relatively large differences between range rate measurements for various portions
Solution Approach 1:
The patent segments the object tracking problem by identifying a specific portion of the object (such as a corner or edge) and focusing measurements on that segment. The sensor selectively tracks a particular portion rather than processing all portions equally, which reduces the impact of range rate variations across different object portions and improves velocity estimate reliability.
Solution Approach 2:
The patent applies local quality by determining that certain portions of the object provide more reliable tracking information than others. The system identifies and prioritizes measurements from specific object portions (e.g., corners or edges with favorable geometry) that have smaller range rate variations, thereby improving the overall measurement precision for velocity estimation.
2Productivity
If noisy sensor measurements corresponding to reflections that bounced off the ground are included in Kalman updates, then more measurements are available for tracking, but the accuracy and certainty in object tracking is reduced
Solution Approach 1:
The patent extracts and removes noisy measurements from the processing pipeline. By identifying measurements with high noise levels (such as those from ground-bounced reflections) and excluding them from Kalman updates, the system maintains tracking productivity while significantly improving measurement precision and tracking accuracy.
Solution Approach 2:
The patent changes the selection criteria for which measurements are processed. Instead of uniformly processing all measurements, the system applies parameter-based filtering to select only those measurements from object portions that meet specific quality thresholds, thereby improving tracking accuracy without sacrificing excessive measurement opportunities.
3Loss of information
If multiple sensor measurements corresponding to different portions of an object are processed, then more data is available for tracking, but computational costs increase and accuracy is reduced due to noise
Solution Approach 1:
The patent applies partial action by processing only a subset of available measurements rather than all measurements. By selectively updating the Kalman filter with measurements from specific object portions that provide the most valuable information, the system reduces computational complexity while maintaining or improving tracking accuracy compared to processing all measurements.
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 enhances tracking accuracy by reducing computational costs and errors associated with traditional clustering techniques, providing more precise velocity and position estimates for objects with multiple portions.
Implementation Method 1
Sensors—such as, by way of example and not limitation, RADAR (RAdio Detection And Ranging) sensors—are often used to track objects.
Implementation Method 2
the tracking of the objects may be performed using a Kalman filter which may be configured to determine and update state estimates of the object based on previous state estimates and sensor data
Data Source
AI summary
One or more embodiments of the present disclosure relate to identifying reference portions corresponding to a bounding shape that corresponds to an object. Additionally, the reference portions may include a first reference edge, a second reference edge, and a reference where the first reference edge and the second reference edge intersect. In some embodiments, operations may further include obtaining a first state estimate corresponding to the object and receiving first sensor data corresponding to a first portion of the object, the first sensor data including a first position measurement. Further, operations may further include determining that the first position measurement corresponds to a first reference portion that is one of the reference portions corresponding to the bounding shape and determining a first expected position corresponding to the first portion based at least on the first reference portion. Embodiments may additionally include determining a second position estimate corresponding to the object.