UAV Multi-Target Tracking Using Outer-Frame Vision Control

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current target tracking technologies for unmanned aerial vehicles (UAVs) are limited to single-target tracking and struggle with efficiently managing multiple targets moving at different speeds and distances, requiring a policy to maximize tracking duration and select targets for optimal performance.

Innovation Solution

A method where the UAV determines and generates an outer frame around multiple targets, adjusts flight parameters to maintain the frame within its vision angle, and implements a target discarding policy to ensure robust multi-target tracking by prioritizing and discarding targets based on identification and distance from the frame edge.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the UAV tracks multiple targets simultaneously, then the tracking coverage and utility are improved, but the device complexity and difficulty of managing different motion statuses increase

Engineering Contradiction:
Improvetracking coverageVSAvoidtracking management complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the multi-target tracking problem into distinct motion status categories (stationary, uniform linear motion, accelerated motion). By classifying targets into these segments, the system applies different tracking policies to each segment, simplifying the overall management complexity while maintaining comprehensive tracking coverage.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements dynamic target discarding policies that adapt based on spatial range and motion status. When the spatial range exceeds a threshold or when tracking duration requirements are met, the system dynamically discards lower-priority targets. This dynamic adjustment reduces tracking management complexity while preserving adaptability to different scenario requirements.

Inventive Principle:
Principle #15Dynamics

2Area of stationary object

If the spatial range of multiple targets is excessively large, then the tracking area coverage is improved, but the tracking precision and performance deteriorate

Engineering Contradiction:
Improvetracking area coverageVSAvoidtracking precision
Core Design Contradiction:
Area of stationary objectVSMeasurement precision

Solution Approach 1:

The patent changes the parameter of spatial range by implementing a threshold-based discarding mechanism. When the spatial range of tracked targets exceeds a predetermined threshold, the system automatically discards certain targets to restore optimal tracking precision. This parameter adjustment maintains the balance between coverage area and tracking precision.

Inventive Principle:
Principle #35Parameter changes

Solution Approach 2:

The patent applies local quality by implementing different tracking precision requirements for different target groups. High-priority targets maintain strict precision requirements, while discarded or lower-priority targets have relaxed requirements. This localized quality approach allows large spatial coverage while maintaining high precision for critical targets.

Inventive Principle:
Principle #3Local quality

3Reliability

If the UAV adjusts flight parameters frequently to maintain all targets in vision range, then the tracking completeness is improved, but the energy consumption increases

Engineering Contradiction:
Improvetracking completenessVSAvoidenergy consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent implements a target discarding mechanism that reduces the number of active tracking targets when spatial range or motion status differences become excessive. By discarding lower-priority targets, the system reduces the frequency of flight parameter adjustments needed to maintain all targets in the vision range, thereby reducing energy consumption while maintaining tracking completeness for high-priority targets.

Inventive Principle:
Principle #34Discarding and recovering

4Device complexity

If the UAV maintains a fixed vision angle, then the device complexity is reduced, but the tracking adaptability to different target configurations deteriorates

Engineering Contradiction:
Improvevision control complexityVSAvoidtracking adaptability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent applies preliminary action by pre-calculating the minimum vision angle required based on target spatial distribution and motion status before executing tracking. This preliminary calculation allows the system to adjust the vision angle proactively rather than reactively, improving tracking adaptability while avoiding excessive complexity in real-time control adjustments.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3474111B1Target tracking method by an unmanned aerial vehicle
Publication Date: 2023.03.29 AUTEL ROBOTICS CO LTD
  • EP3474111B1 patent drawingFigure 1~2
  • EP3474111B1 patent drawingFigure 3
  • EP3474111B1 patent drawingFigure 4

AI summary

The embodiments of the present invention disclose a target tracking method, an unmanned aerial vehicle, and a computer readable storage medium. The method includes: determining, by an unmanned aerial vehicle, at least two targets that need to be tracked; generating, by the unmanned aerial vehicle, an outer frame according to location information of the at least two targets, where the at least two targets are located within the outer frame or on the outer frame; and adjusting, by the unmanned aerial vehicle, a flight parameter of the unmanned aerial vehicle, so that the outer frame falls within a vision angle range of the unmanned aerial vehicle. According to the solutions of the embodiments of the present invention, the unmanned aerial vehicle generates the outer frame according to the location information of the at least two targets, and the unmanned aerial vehicle adjusts the flight parameter of the unmanned aerial vehicle, so that the outer frame falls within the vision angle range of the unmanned aerial vehicle. In this way, multiple targets are converted to a "single target", to find an optimal location for observing all the multiple targets, and it is ensured that the multiple targets are tracked with robustness for maximum duration.