Search papers, labs, and topics across Lattice.
This paper introduces a data-driven approach for enhancing the autonomy of micro-UAVs in GPS-denied environments, specifically targeting disaster response scenarios. By utilizing a custom micro-drone equipped with a 360-degree camera, the authors recorded diverse scenarios to train convolutional neural networks for predicting yaw commands from monocular views. The results indicate that the learned yaw controller can effectively navigate micro-drones toward open spaces, preserving critical sensor data even in communication-limited conditions, although challenges such as reflections and glare were identified during real-world testing.
Micro-drones can autonomously navigate hazardous environments without GPS, preserving vital sensor data even when communication is lost.
This paper presents a novel data-driven approach to camera-based autonomy for micro-drones in GPS-denied, radio-challenging indoor environments. The target application is disaster and emergency response, where micro-UAVs can provide rapid situational awareness in hazardous settings such as firefighting and chemical, biological, radiological, and nuclear (CBRN) incidents while reducing risk for human responders. When the communication link is lost, the micro-drone uses a learned yaw controller to autonomously navigate toward open space, preserving onboard sensor data that would otherwise be lost with the vehicle. A custom micro-drone equipped with a 360-degree camera was used to record diverse industrial, underground, and training scenarios representative of communication-denied field operations. We introduce a preprocessing pipeline that converts equirectangular 360-degree footage into planar front views and dynamically generates image-label pairs for AI training. We then train and compare multiple convolutional neural network variants that predict a continuous yaw command from a single monocular view. Evaluation on a held-out test set confirms the feasibility of the learned yaw-prediction approach. A semi-autonomous real-world test further demonstrates the practicality of the method while revealing key failure modes, particularly reflections and glare.
