Unlock instant, AI-driven research and patent intelligence for your innovation.

Multi-target tracking positioning and motion state estimation method based on unmanned aerial vehicle

A multi-target tracking and motion state technology, applied in neural learning methods, calculations, computer components, etc., to achieve precise calculations and improve accuracy

Active Publication Date: 2021-08-17
NAT INNOVATION INST OF DEFENSE TECH PLA ACAD OF MILITARY SCI
View PDF5 Cites 26 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Benefits of technology

This technology allows for precise navigation over long distances by creating an avatar's visual representation based upon data from multiple cameras mounted at different locations around it. It uses advanced techniques like image processing algorithms or machine learning methods to improve its performance. Overall, this technology helps navigate through space without being obstructive even when there are many objects within sight (such as buildings).

Problems solved by technology

Technological Problem addressed in this patented technical solution describes how complex object detection systems like Multi Object Tracking (MOFT), which involves multiple objects being tracked together over different terrains or areas covered by obstacles during flight. These challenges include environmental disturbances caused by sunlight glare, shading, movement within buildings, and other sources affecting visual quality. Current algorithms have limitations including slow processing times due to large computational costs associated with these techniques.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Multi-target tracking positioning and motion state estimation method based on unmanned aerial vehicle
  • Multi-target tracking positioning and motion state estimation method based on unmanned aerial vehicle
  • Multi-target tracking positioning and motion state estimation method based on unmanned aerial vehicle

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0089] In order to better understand the contents of the present invention, an example is given here.

[0090] figure 1 It is the tracking flow chart of the present invention, and the present invention discloses a multi-target tracking and positioning and motion state estimation method based on unmanned aerial vehicles, and its specific steps include:

[0091] S1. Obtain the target observation image data from the aerial view of the UAV, use YOLOv4 as the detector, detect the target bounding box bbox in the current frame, convert the detected target bounding box bbox into the target detection result, and obtain the target detection result , including detection frame, target category, confidence and other information. In the UAV scenario, the target detection algorithm enables the UAV to quickly detect vehicles and pedestrians on the ground to make real-time decisions.

[0092] Step S1 specifically includes,

[0093] S11, initialize each parameter of the target detection resu...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a multi-target tracking positioning and motion state estimation method based on an unmanned aerial vehicle, and the method comprises the specific steps: inputting target observation image data under the aerial photographing of the unmanned aerial vehicle at a high-altitude visual angle, and obtaining a target detection result; predicting target trajectory parameters by using Kalman filtering, then carrying out cascade matching by using a Hungary algorithm, carrying out IoU matching on a cascade matching result, updating the state of a target trajectory by using Kalman filtering, updating each successfully matched trajectory by using a corresponding detection result, and processing the unmatched trajectories and unmatched detection results; and transplanting the result into an ROS environment, converting pixel coordinates on a two-dimensional image shot by a camera of the unmanned aerial vehicle into space coordinates of the real world, and calculating the moving speed of the target. According to the method, migration adaptation of a multi-target tracking and positioning algorithm is completed in the operation process of the high-altitude unmanned aerial vehicle, and accurate calculation of the multi-target motion state is achieved.

Description

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Owner NAT INNOVATION INST OF DEFENSE TECH PLA ACAD OF MILITARY SCI
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More