Drone Landing on Moving Base Using Depth Sensor and Computer Vision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in safely and precisely landing drones on moving bases, such as car drone docking stations, especially in busy traffic conditions, which limits their flight duration and range, and hinders large-scale commercialization.
Innovation Solution
The integration of a depth sensor and a processor configured to control a computer vision tracking algorithm, converting RGB frames to HSV frames, computing a saliency score, and performing frame smoothing to enhance tracking accuracy and jitter control, allowing drones to land smoothly on moving bases by adjusting their movement based on distance and obstacle detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Duration of action of moving object
If a drone attempts to land on a moving base in busy traffic conditions, then the flight duration and range can be extended, but the landing safety and precision deteriorate
Solution Approach 1:
The system employs computer vision tracking algorithms that continuously monitor the relative position and motion of the moving base, providing real-time feedback to adjust drone trajectory and landing姿态. The depth sensor data feeds back into the control system to dynamically modify descent rate and horizontal positioning, enabling safe landing despite platform motion and environmental distractions.
Solution Approach 2:
The patent introduces an intermediary processing layer between the raw sensor data and the control actuators. The computer vision system acts as an intermediary that processes visual information about the moving base and translates it into actionable tracking data. This intermediary layer filters out irrelevant visual noise from busy environments while extracting critical motion parameters needed for safe landing.
2Measurement precision
If computer vision tracking is increased when distance is within threshold, then tracking accuracy improves, but computational load and processing time increase
Solution Approach 1:
The system dynamically adjusts the intensity of computer vision tracking based on real-time distance measurements from the depth sensor. When the drone is far from the moving base, standard tracking is sufficient. As the drone approaches and enters the distance threshold, the system automatically increases tracking accuracy and processing frequency. This dynamic adaptation ensures high precision when needed while minimizing computational overhead during earlier flight phases.
Solution Approach 2:
The patent changes key processing parameters based on distance conditions. The frame rate, resolution, and algorithmic complexity of the computer vision system are adjusted as functions of distance to the moving base. Closer distances trigger higher frame rates and more intensive processing, while greater distances allow for reduced processing intensity. This parameter adaptation resolves the contradiction between accuracy and computational load.
3Stability of the object's composition
If frame conversion from RGB to HSV is performed to control jitter, then tracking stability improves, but processing time increases
Solution Approach 1:
The system applies frame conversion from RGB to HSV color space periodically rather than continuously, specifically during phases when jitter control is most critical such as during final approach and landing. The conversion is triggered by distance thresholds and motion state detection, converting frames only when the moving base or drone exhibits significant vibration or when precision tracking is paramount. This periodic application reduces overall processing time while maintaining tracking stability when most needed.
Data Source
AI summary
A drone is described. The drone includes a depth sensor configured to provide information for determining a distance between the drone and a moving base. The drone also includes a processor configured to control a computer vision tracking algorithm based on the distance, and to control drone movement based on the computer vision tracking algorithm. A vehicle is also described. The vehicle includes a depth sensor configured to provide information for determining a distance between a drone and the vehicle. The vehicle also includes a processor configured to control a computer vision tracking algorithm based on the distance and to send information for controlling drone movement based on the computer vision tracking algorithm.