UAV External State Estimation with Monocular Vision and Range Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unmanned aerial vehicles (UAVs) using visual navigation experience a deviation in autonomously determined state information over time, reducing the reliability of their navigation systems.
Innovation Solution
The implementation of a method and system that updates state information by calculating a relative proportional relationship between a camera-based reference frame and a global reference frame using a monocular camera unit and a proximity sensor, such as lidar, ultrasound, or infrared sensors, to improve accuracy and reliability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If visual navigation is used for autonomous UAV flight, then the UAV can navigate without human guidance, but the determined state information deviates from actual state over time, reducing navigation reliability
Solution Approach 1:
The patent combines visual navigation with ultrasonic ranging technology to create a hybrid navigation system. The ultrasonic sensor provides accurate distance measurements that are fused with visual navigation data, correcting drift accumulation and improving overall navigation reliability while maintaining autonomous operation
Solution Approach 2:
The system implements feedback by continuously comparing ultrasonic range measurements with visually determined state information. This feedback loop detects deviations between actual and estimated states, enabling real-time correction of navigation errors and maintaining reliability over extended flight periods
2Device complexity
If monocular camera is used for state determination, then the system structure remains simple, but measurement precision of external state information deteriorates due to scale ambiguity
Solution Approach 1:
The ultrasonic sensor acts as an intermediary that provides absolute distance measurements to resolve the scale ambiguity inherent in monocular vision. By using ultrasonic range data as a reference, the system can determine the actual scale of visual features, thereby improving measurement precision without significantly increasing system complexity
Solution Approach 2:
The patent changes the parameter of distance measurement by introducing ultrasonic ranging capability. This additional measurement parameter enables the system to determine absolute scale and depth information that complements the relative measurements from the monocular camera, improving overall state estimation accuracy
Data Source
AI summary
A method for determining an external state of an unmanned aerial vehicle (UAV) includes obtaining historical external state information of the UAV, obtaining a current image and a historical image captured before the current image, predicting a matching feature point in the current image that corresponds to a target feature point in the historical image according the historical external state information, and determining the external state of the UAV based on the matching feature point.