Vision-Based Autonomous UGV Detection, Tracking, and Following for a UAV

What is it about?

This study introduces a method for guiding unmanned ground vehicles (UGVs) in environments where traditional GPS navigation is ineffective, such as dense forests or rugged terrains. Our approach leverages unmanned aerial vehicles (UAVs) to visually monitor and direct the UGVs, enabling them to navigate through challenging landscapes. The process begins with the UAV capturing aerial imagery to locate the UGV. Subsequently, the UAV continuously tracks the UGV, determining its precise location and orientation. This tracking allows the UAV to maintain a consistent position above the UGV, facilitating continuous guidance. In addition to real-time monitoring, the UAV transmits the captured images to a base station, where the data is used to generate a global traversability map. This map categorizes the terrain into navigable and non-navigable areas, guiding the UGV's movement. To enhance the accuracy of this map, we propose integrating additional sensors, such as LiDAR, radar, and inertial measurement units (IMUs), into the system. Currently, our research focuses on the initial stages of UAV-based detection, tracking, and relative positioning of the UGV. Future work will delve into the advanced development of the global traversability map and its application in autonomous UGV navigation. This UAV-UGV collaboration presents a novel solution to the challenge of autonomous navigation in GPS-denied environments, offering significant potential for applications in areas where human access is restricted or dangerous, including disaster relief, military reconnaissance, and environmental monitoring.

Featured Image

Photo by Jared Brashier on Unsplash

Photo by Jared Brashier on Unsplash

Why is it important?

This UAV-UGV collaboration offers a timely solution to the limitations faced by ground-based autonomous systems. By leveraging aerial perspectives for navigation, our methodology extends the operational capabilities of UGVs beyond the constraints of their on-board sensors and the unreliable GPS signals in certain areas. This opens up new possibilities for the deployment of unmanned systems in areas previously considered inaccessible or too risky for autonomous navigation.

Perspectives