Object Tracking Aliased Velocity Selection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing object tracking systems face challenges in accurately identifying the true relative velocity of objects due to ambiguity in phase rotation of frequency components, leading to increased processing load and potential object recognition failures when reducing the number of velocity assumptions.
Innovation Solution
An object tracking apparatus that calculates a plurality of aliased velocities and generates target candidates, selecting the true target based on observation information, including distance, relative velocity, and orientation, while setting a velocity lower-limit value based on the moving body's velocity to improve estimation accuracy and reduce processing load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If multiple velocity assumptions are used to track objects, then object recognition reliability is improved, but processing load increases
Solution Approach 1:
The velocity range is segmented into multiple discrete velocity assumptions (e.g., -30 m/s to +30 m/s in 5 m/s increments), allowing the system to divide the continuous velocity space into manageable discrete options for tracking different object types
Solution Approach 2:
The system generates target candidates for all velocity assumptions simultaneously (excessive action), then uses observation information to eliminate false candidates, rather than trying to determine the correct velocity assumption in advance
2Measurement precision
If velocity lower-limit value is set based on moving body velocity, then estimation accuracy of true relative velocity is improved, but device complexity increases
Solution Approach 1:
The velocity lower-limit value is pre-calculated based on the moving body's velocity before target detection begins, establishing a foundation for generating appropriate target candidates and reducing the search space for true targets
Solution Approach 2:
The system dynamically adjusts the velocity lower-limit value parameter based on the moving body's velocity, which changes the range of aliased velocities that need to be considered for target candidate generation
Data Source
AI summary
An object tracking apparatus acquires detection information from a sensor mounted to a moving body, detects an object, and calculates a distance, a relative velocity, and an orientation thereof. For the initially detected object, the apparatus calculates aliased velocities of which aliasing from the relative velocity is assumed, and generates target candidates corresponding to the aliased velocities. For each target candidate, the apparatus estimates a current state of the target candidate from a past state thereof and observation information, and selects the target candidate estimated to be a true target from the target candidates for the same object. The apparatus calculates, as a reference velocity, an aliased velocity of the relative velocity that is equal to or greater than a velocity lower-limit value and is most negative or smallest in magnitude. The apparatus calculates the aliased velocities being the reference velocity aliased 0 to n times in a positive direction.