UAV Body-Pose Control Using Onboard Vision Sensing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Users face frustration and inefficiency when operating unmanned aerial vehicles (UAVs) due to the need for separate devices and attention diversion, limiting their ability to fully utilize UAV intelligence and capture timely subject matter with on-board cameras.
Innovation Solution
A method and apparatus that utilize image data from an on-board imaging sensor to identify human body indications, allowing UAVs to operate in response, thereby enabling users to control the device through natural body poses and movements without the need for additional devices.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a secondary device (controller or mobile phone) is used to operate the UAV, then the UAV can be controlled with precise commands, but the user requires extra effort and time to learn and practice the controlling process
Solution Approach 1:
The patent replaces the mechanical control system (physical controller with joysticks and buttons) with a vision-based control system. The UAV's imaging sensor captures images of the user's body, and image processing algorithms automatically interpret body poses and movements to generate control commands, eliminating the need for physical controllers and reducing the learning curve.
Solution Approach 2:
The system enables self-service control where the user's natural body movements are automatically captured and interpreted by the UAV's onboard imaging sensor and processing system. The UAV autonomously tracks the user's body pose changes and converts them into control commands without requiring manual intervention through a controller.
2Measurement precision
If the user transfers attention to operate the controller or mobile phone, then precise control commands can be sent to the UAV, but the user gets distracted from the ongoing activity
Solution Approach 1:
The patent replaces manual controller operation with automatic vision-based control. The imaging sensor continuously captures the user's body pose, and the system automatically processes these images to generate control commands, allowing the user to remain engaged in their activity while the UAV is controlled through passive body movement tracking.
Solution Approach 2:
The system maintains continuous control of the UAV through ongoing image capture and processing of the user's body pose. Unlike intermittent controller input, the imaging sensor continuously tracks the user's movements, providing uninterrupted control signals to the UAV while the user remains engaged in their primary activity.
3Reliability
If conventional control methods are used, then the UAV can be operated with dedicated control devices, but users are frustrated by the cumbersome experience and discouraged from using UAVs
Solution Approach 1:
The patent replaces complex mechanical control interfaces with intuitive vision-based control. Natural body poses and movements are captured by the imaging sensor and automatically translated into reliable control commands, maintaining control reliability while dramatically improving ease of operation and user satisfaction.
4Measurement precision
If the user focuses on controller operation, then control commands can be executed accurately, but the user misses opportunities to timely record subject matter of interest
Solution Approach 1:
The system enables the UAV to autonomously respond to the user's body movements and capture subject matter of interest. When the user naturally points or gestures toward a subject, the system automatically interprets this body language and directs the UAV to record the target, eliminating the need for the user to manually switch between controlling and capturing.
Data Source
AI summary
Methods, apparatuses, and non-transitory computer-readable medium for operating a movable object, the method including obtaining image data based on one or more images captured by an imaging sensor on board the movable object. Each of the one or more images includes at least a portion of a first human body. The method also includes identifying a first indication of the first human body in a field of view of the imaging sensor based on the image data, and causing the movable object to operate in response to the identified first indication of the first human body in the field of view of the imaging sensor.