UAV PTZ Target Position Estimation for Stable Long-Range Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing position estimation methods for tracking targets using unmanned aerial vehicles (UAVs) are inaccurate and unstable, particularly at long distances, due to reliance on binocular cameras with limited field of view and ground altitude information that degrades over time.
Innovation Solution
A position estimation method using a pan-tilt-zoom (PTZ) camera to model the tracking target as a sphere, determine its radius, and calculate height and width differences in images, combined with an extended Kalman filter (EKF) algorithm to update target positions based on these differences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If binocular camera back-projection method is used, then measurement precision is improved, but field of view coverage deteriorates
Solution Approach 1:
The patent divides the tracking system into two segments: binocular camera for close-range high-precision measurement and PTZ camera for far-range tracking. Each segment operates in its optimal range, with the system switching between them based on distance thresholds, thus resolving the contradiction between measurement precision and field of view coverage
Solution Approach 2:
The patent introduces a PTZ camera as an intermediary device to extend the field of view coverage. The PTZ camera captures wide-area images that are then processed to estimate target positions for targets outside the binocular camera's field of view, acting as a mediator between the limited field of view and the need for comprehensive coverage
2Device complexity
If ground altitude information is used for position estimation, then device complexity is reduced, but measurement precision deteriorates over time
Solution Approach 1:
The patent implements feedback by using the PTZ camera to continuously monitor and estimate target positions, providing corrective information to compensate for the drift in ground altitude data. The system compares estimated positions with actual tracking positions and adjusts accordingly, maintaining precision without increasing overall system complexity
Solution Approach 2:
The patent replaces reliance on mechanical inertial measurement units (IMU) and ground altitude sensors with an optical-based PTZ camera system for position estimation. This substitution uses visual information processing instead of mechanical sensing, eliminating the accumulation error inherent in mechanical systems while keeping device complexity manageable
3Device complexity
If PTZ camera pitch angle is used for distance calculation, then device complexity is reduced, but measurement precision deteriorates at long distances
Solution Approach 1:
The patent merges multiple measurement approaches: it combines PTZ camera pitch angle data with image processing algorithms that analyze target size and position in the captured image. By merging these methods, the system compensates for the limited sensitivity of pitch angle measurements at long distances while maintaining relatively simple device complexity
Solution Approach 2:
The patent transitions from relying solely on one-dimensional pitch angle measurements to utilizing two-dimensional image space information. By analyzing the target's position, size, and shape in the image plane alongside pitch angle data, the system gains additional dimensional information that improves distance measurement precision without significantly increasing device complexity
Data Source
AI summary
A position estimation method for a tracking target is implemented in an unmanned aerial vehicle. The position estimation method include: estimating a target position of the tracking target at the next time according to an initial position of the tracking target at the current moment; determining an estimated width and an estimated height of the tracking target in an image captured by a pan-tilt-zoom camera of the unmanned aerial vehicle according to the estimated target position; obtaining an actual width and an actual height of the tracking target in the image; determining a height difference between the estimated width and the estimated height and a width difference between the actual height and the actual width; and updating the target position of the tracking target at the next time according to the height difference and the width difference.