Chinese researchers have put forth a new method to manage unmanned aerial vehicles (UAVs) in situations where GLONASS or GPS signals are not available. The method, called “Image-Based Visual Serving” (IBVS), utilizes an observer of speed for UAVs to track dynamic objects in the absence of GLONASS or GPS.
Unlike existing methods, the proposed approach uses simplified and declared image dynamics for UAVs with the help of a virtual chamber. This accounts for uncertainty caused by unpredictable rotations and speeds of dynamic targets.
Researchers have also introduced a new model of image depth that enables the IBVS method to monitor rotating targets with various orientations. This model ensures accuracy of image features and trajectory smoothness during tracking.
The use of an observer is a key aspect of the method. It eliminates the need for speed measurement and reduces interference caused by noisy dimensions.
To compensate for unpredictable external interference, the researchers have suggested the use of an integral filter. The stability of the observer of speed and controller of IBVS has been analyzed using the Lyapunov method.
Comparative modeling and multi-stage experiments have demonstrated the tracking stability, interference resistance, and stability of the proposed method when tracking dynamic rotating goals.
Although the basic principles of this technique have been known since the 1980s, the innovation lies in the use of advanced algorithms and sensors, enabling robots to effectively “see” and perform complex tasks.
The new functionality is expected to be implemented in military UAVs, especially in light of the increasing electronic warfare. However, human intervention is still usually required for the final decision on an attack.
The study can be found in the Engineering magazine.