Gesture Control of Micro-drone: A Lightweight-Net with Domain Randomization and Trajectory Generators
This work addresses the challenge of intuitive, computationally-efficient control for micro-drones in industrial applications, representing an incremental improvement over existing techniques.
The study tackled the problem of controlling micro-drones with limited computing power by developing a lightweight deep convolutional neural network using Gabor filters and spatial separable convolutions, which conserved nearly 18% of computational resources during training and showed promising results in gesture-based control compared to state-of-the-art methods.
Micro-drones can be integrated into various industrial applications but are constrained by their computing power and expert pilots, a secondary challenge. This study presents a computationally-efficient deep convolutional neural network that utilizes Gabor filters and spatial separable convolutions with low computational complexities. An attention module is integrated with the model to complement the performance. Further, perception-based action space and trajectory generators are integrated with the model's predictions for intuitive navigation. The computationally-efficient model aids a human operator in controlling a micro-drone via gestures. Nearly 18% of computational resources are conserved using the NVIDIA GPU profiler during training. Using a low-cost DJI Tello drone for experiment verification, the computationally-efficient model shows promising results compared to a state-of-the-art and conventional computer vision-based technique.