UAV Navigation via Remote 3D Virtual Model
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Unmanned aerial vehicles (UAVs) face challenges in indoor environments due to the need for precise control to prevent accidents and the limitations of lightweight drones, which cannot carry large sensing systems or powerful processing systems required for collision prevention, especially in industrial settings with moving objects.
Innovation Solution
A computer-implemented method and system for controlling UAVs that enables autonomous navigation and task execution, such as product searching and obstacles identification, using lightweight drones equipped with a flight controller, inertial measuring units, and imaging sensors, with a remote computing server creating a 3D virtual model of the environment for precise control and obstacle avoidance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Weight of moving object
If lightweight drones are used to operate in indoor environments with human beings, then the drones can safely coexist with humans in the same space, but they are unable to carry large sensing systems and powerful processing systems required for collision prevention
Solution Approach 1:
The patent introduces an external computing server as an intermediary to handle the complex sensing and processing tasks. The lightweight drone carries only minimal onboard equipment (IMU, basic communications), while the computing server receives data from the drone and performs the heavy computational work for creating 3D virtual models and calculating collision-free routes, thereby resolving the contradiction between drone weight and collision prevention capability
Solution Approach 2:
The patent replaces the traditional mechanical approach of equipping drones with heavy onboard sensors and processors with an information-based system. Instead of mechanically carrying all necessary computing resources, the drone uses wireless communication to transmit data to an external server that performs the computational functions, substituting physical mechanical capacity with information transmission and remote processing
2Reliability
If precise control systems are implemented to prevent accidents in indoor environments, then safety is improved, but the complexity of the control system increases
Solution Approach 1:
The computing server acts as an intermediary that centralizes the complex control logic. Instead of distributing complex control algorithms across the drone itself, the server receives simple telemetry data from the drone and returns sophisticated control commands, thereby achieving precise safety control without increasing onboard system complexity
Solution Approach 2:
The patent creates a 3D virtual model (digital twin) of the physical environment as a copy. This virtual replica allows the system to simulate and calculate safe routes in the virtual space before executing them in the physical world, enabling precise control decisions without requiring complex real-time sensing and processing on the drone
3Extent of automation
If autonomous navigation capabilities are added to drones for complex tasks, then automation level increases, but the processing power requirements exceed what lightweight drones can carry
Solution Approach 1:
The computing server serves as an external brain that provides autonomous navigation capabilities. The lightweight drone maintains only basic autonomous functions (flight stabilization via IMU, wireless communication), while the server performs advanced autonomous tasks such as real-time 3D mapping, obstacle detection, and dynamic route planning, achieving high-level automation without increasing drone weight
Solution Approach 2:
The patent moves the processing dimension from the physical space (onboard the drone) to the cyber space (remote server). By transitioning computational resources from the physical drone platform to a remote digital infrastructure, the system achieves powerful autonomous navigation capabilities while keeping the physical drone lightweight
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer implemented method of controlling navigation of at least one drone, is disclosed. The method comprises: from each drone navigating in a facility (10), and equipped with an imaging sensor, sending a plurality of pictures to a server (30) through a wireless communications interface (152); at the server (30): receiving the pictures and obtaining a 3D virtual model of the facility (10) in which the drone is navigating, the 3D virtual model comprising, for each drone: a first virtual drone located within the 3D virtual model at a position and orientation calculated from the received pictures, said position and orientation representing the position and orientation of the drone; and a second virtual drone located within the 3D virtual model at a target position and orientation representing the position and orientation to be reached by the drone; for each drone, sending to the drone at least one movement command (416, 417) for reaching the point at the facility (10) which corresponds to the location of the second virtual drone at the 3D virtual model; the at least one drone navigating according to the at least one movement command (416, 417) and, from the new location of the at least one drone (15, 15_1, ... 15_N), sending at least one picture captured by the imaging sensor (156) to the server (30); at the server (30): from information extracted from the pictures sent by each drone, updating the 3D virtual model of the facility (10) and the location of each first virtual drone in the 3D virtual model, in such a way that changes in the facility (10) captured by the at least one picture are reflected in the 3D virtual model; and repeating the former steps until the drone reaches the position and orientation of the second virtual drone as defined in the 3D virtual model.