Vision-Based Drone Navigation Transition After Takeoff
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous drones face challenges in transitioning from initial navigation after takeoff to vision-based navigation due to size and power constraints, especially when GPS is unavailable or inoperative, requiring a reliable method to stabilize and safely navigate using image analysis.
Innovation Solution
The drone initiates by capturing an image of a face or person to determine distance and stabilize, then hovers to gather sensor data, transitioning to a second navigation module when sufficient data is collected, allowing it to take over navigation to waypoints while capturing images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the drone uses GPS-based navigation, then navigation reliability is improved, but the drone cannot operate indoors or in GPS-denied environments
Solution Approach 1:
The navigation system is segmented into multiple independent modules: a first navigation module for initial stabilization and a second vision-based navigation module for waypoint navigation. This segmentation allows each module to specialize in specific tasks, enabling reliable operation in GPS-denied environments by combining traditional navigation with vision-based navigation.
Solution Approach 2:
The drone's navigation system is designed with multi-functionality to operate in both GPS-available and GPS-denied environments. The system can switch between GPS-based navigation and vision-based navigation, making it universally applicable across different operational contexts including indoor and outdoor environments.
2Adaptability or versatility
If the drone transitions to vision-based navigation immediately after takeoff, then navigation adaptability is improved, but the drone lacks sufficient sensor data for stable navigation
Solution Approach 1:
The first navigation module performs preliminary stabilization actions immediately after takeoff by capturing images of the operator's face and using facial recognition to maintain a stable hover. This preliminary stabilization ensures the drone is properly positioned and stabilized before transitioning to the second vision-based navigation module, providing a reliable foundation for subsequent waypoint navigation.
Solution Approach 2:
The first navigation module acts as an intermediary between takeoff and full vision-based navigation. It captures initial sensor data and stabilizes the drone, serving as a bridge that prepares the system for the second navigation module. This intermediary phase ensures sufficient sensor data is collected before transitioning to autonomous waypoint navigation.
3Reliability
If the drone collects sufficient sensor data before transitioning to vision-based navigation, then navigation stability is improved, but the transition time increases
Solution Approach 1:
The system collects just enough sensor data during the preliminary stabilization phase to ensure stable navigation, rather than collecting excessive data. The first navigation module captures images and sensor data during the hover phase, which is sufficient for the second vision-based navigation module to take over. This partial action approach balances data collection needs with time efficiency.
4Ease of operation
If the drone uses a light, portable body, then ease of operation is improved, but power constraints limit navigation capabilities
Solution Approach 1:
The navigation system is segmented into two phases with different power requirements. The first navigation module handles initial stabilization with lower power consumption, while the second vision-based navigation module handles waypoint navigation. This segmentation allows the lightweight drone to operate within power constraints by distributing navigation tasks across different modules.
Solution Approach 2:
The drone performs periodic actions: first stabilizing and hovering to collect sensor data, then transitioning to waypoint navigation. This periodic pattern of stabilization followed by navigation allows the system to manage power consumption efficiently, alternating between data collection phases and execution phases rather than continuously consuming high power.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems, computer readable medium and methods for autonomous drone navigation based on vision are disclosed. Example methods include capturing an image using an image capturing device of the autonomous drone, processing the image to identify an object, and navigating the autonomous drone relative to the object for a period of time. After the period of time a second type of navigation is used based on determining structure from motion navigation. Images are captured during the period of time to transition to the second type of navigation. The second type of navigation uses a downward pointing navigation camera and other sensors.