Drone Flight Planning for High-Resolution Road Imagery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Aerial imagery captured by human-operated aircrafts lacks high resolution and fails to capture necessary road features for autonomous navigation, while on-vehicle cameras can only capture partial intersections with inaccurate lane widths and cut-off road curvatures.
Innovation Solution
A reinforcement learning-based system controls a drone to follow a target vehicle, generating dynamic flight plans and maintaining line of sight to capture high-resolution, relevant road features with reduced unwanted imagery, aligning aerial and ground data for precise feature extraction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If human-operated aircrafts are used to capture aerial imagery, then coverage area is large, but image resolution is low and road features are not captured clearly
Solution Approach 1:
The system transitions from fixed-altitude aerial imaging to three-dimensional dynamic following, where the drone moves vertically and horizontally to maintain optimal positioning behind the target vehicle, capturing high-resolution imagery from multiple angles and distances
Solution Approach 2:
The drone acts as an intermediary between human-operated aircraft and ground-level cameras, providing aerial perspective while maintaining close proximity to the target vehicle through active tracking, thus achieving both coverage and resolution
2Measurement precision
If on-vehicle cameras are used to capture road features, then image resolution is high, but field of view is limited and road curvatures are cut off
Solution Approach 1:
The system moves from ground-level two-dimensional imaging to aerial three-dimensional imaging, where the drone's elevated position provides an expanded field of view that captures road curvatures, intersections, and surrounding context while maintaining high resolution through proximity to the target vehicle
Solution Approach 2:
The drone dynamically adjusts its position, altitude, and orientation in real-time to follow the target vehicle, expanding and shifting the field of view to capture road features ahead of, beside, and behind the vehicle that would be invisible to fixed on-vehicle cameras
3Area of stationary object
If aerial imagery is captured from high altitude, then coverage area is large, but unwanted surrounding imagery increases data processing needs
Solution Approach 1:
The system extracts and isolates only the relevant road features and surrounding context needed for autonomous navigation by having the drone follow the target vehicle and capture imagery focused on the road ahead, sides, and immediate surroundings, excluding irrelevant distant areas
Solution Approach 2:
The system applies different imaging qualities and focus levels to different spatial zones, with high-resolution detailed capture of road features in the immediate vicinity and selective lower-resolution or excluded capture of distant surrounding areas that are not relevant to navigation
Data Source
AI summary
A method of surveying roads includes generating a dynamic flight plan for a drone using a vehicle traveling on a road as a target. The dynamic flight plan includes instructions for movement of the drone. The method includes controlling the drone as a function of position of the vehicle based on the dynamic flight plan. The method includes maintaining, based on the controlling, line of sight with the drone while the drone with an onboard camera follows the vehicle and captures images of the road being traveled by the vehicle using the onboard camera.