External Object Recognition With Region-Based Tracking Error Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing collision damage mitigation brake systems face challenges in accurately tracking moving objects due to observation errors and detection failures, particularly when using image-based tracking methods, which can lead to incorrect distance measurements and potential collisions.
Innovation Solution
An external recognition device that includes an object position detection unit, region determination unit, observation error setting unit, state prediction unit, and state update unit to enhance the tracking of moving objects by calculating and updating their states based on camera images, using a Kalman filter and error correction techniques.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image-based tracking method is used to detect moving objects, then the device can capture visual information of the moving object, but the observation error varies depending on detection position and tracking accuracy deteriorates
Solution Approach 1:
The patent applies local quality by setting different observation errors for different regions in the image. Specifically, the image is divided into multiple regions, and each region is assigned a different observation error value based on its distance from the camera. Regions closer to the camera (where detection is more reliable) have smaller observation errors, while regions farther away have larger observation errors. This resolves the contradiction by making the measurement precision adaptive to the local detection reliability in each image region.
2Measurement precision
If Kalman filter is used to predict moving object position, then the system can estimate future position based on historical data, but the prediction accuracy reduces when observation error cannot be accurately set
Solution Approach 1:
The patent applies parameter changes by dynamically adjusting the observation error parameter based on the spatial region of detection. Instead of using a fixed observation error value, the system changes the parameter according to the position in the image where the object is detected. This allows the Kalman filter to adapt to varying detection conditions without requiring complex manual error setting, thereby maintaining prediction accuracy while simplifying the overall system complexity.
3Adaptability or versatility
If camera angle and mounting angle are optimized to expand viewing field, then more moving objects can be detected, but detection accuracy decreases at image boundaries
Solution Approach 1:
The patent addresses this contradiction by applying local quality through region-dependent observation error setting. When the camera's viewing field is expanded to cover more areas, the system compensates for reduced detection accuracy at image boundaries by assigning larger observation errors to those boundary regions. This allows the system to maintain broad adaptability and coverage while accurately reflecting the reduced precision at the edges through appropriate error weighting in the tracking algorithm.
Data Source
AI summary
An external recognition device that detects and tracks a moving object includes an object position detection unit that detects a position of the moving object as an observation value, based on an image; a region determination unit that determines a region to which the moving object belongs in the image, based on the observation value; an observation error setting unit that calculates an error to an observation value, based on a determination result; a state prediction unit that predicts a state of the moving object at a current time, based on the observation value up to a previous time that is a time earlier than the current time and the error; an association unit that associates the state of the moving object at the current time with the observation value; and a state update unit that updates the state of the moving object, based on a result of the association.