Autonomous UAV Trail Navigation with Obstacle Avoidance Using Deep Neural Networks
- PDF / 13,119,955 Bytes
- 17 Pages / 595.224 x 790.955 pts Page_size
- 74 Downloads / 212 Views
Autonomous UAV Trail Navigation with Obstacle Avoidance Using Deep Neural Networks Seungho Back1 · Gangik Cho2 · Jinwoo Oh2 · Xuan-Toa Tran3 · Hyondong Oh2 Received: 5 March 2020 / Accepted: 1 September 2020 © Springer Nature B.V. 2020
Abstract This paper proposes a vision-based bike trail following approach with obstacle avoidance using CNN (Convolutional Neural Network) for the UAV (Unmanned Aerial Vehicle). The UAV is controlled to follow a given trail while keeping its position near the center of the trail using the CNN. Also, to return to the original path when the UAV goes out of the path or the camera misses the trail due to disturbances such as wind, the control commands from the CNN are stored for a certain duration of time and used for recovering from such disturbances. To avoid obstacles during the trail navigation, the optical flow computed with another CNN is used to determine the safe maneuver. By combining these methods of i) trail following, ii) disturbance recovery, and iii) obstacle avoidance, the UAV deals with various situations encountered when traveling on the trail. The feasibility and performance of the proposed approach are verified through realistic simulations and flight experiments in real-world environments. Keywords Autonomous navigation · Obstacle avoidance · Deep learning · Trail following · Unmanned aerial vehicle
1 Introduction The applications using UAVs (Unmanned Aerial Vehicles) have grown rapidly in recent years and are widely used for military and civilian purposes. Among them, autonomous navigation of UAVs is one of the active research areas as it Hyondong Oh
[email protected] Seungho Back [email protected] Gangik Cho [email protected] Jinwoo Oh [email protected] Xuan-Toa Tran [email protected] 1
NearthLab, Seoul 06246, South Korea
2
Department of Mechanical Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan 44919, South Korea
3
NTT Hi-Tech Institute, Nguyen Tat Thanh University, 300A Nguyen Tat Thanh Street, Ho Chi Minh City, Viet Nam
can be applied to various fields such as search and rescue, mapping, monitoring, and personal video shooting among many others. In particular, unlike the UGV (Unmanned Ground Vehicle), the UAV has advantages of having much higher speed and functionality without being limited by the shape of the terrain, such as the height of the terrain or the wetland. For autonomous flights, the UAV should fly in unknown environments safely, and the walkway or bicycle path is the one of environments that the UAV can fly long distances with minimal risk using visual cues therein. For an autonomous flight of the UAV, it is necessary to recognize the trail. However, conventional algorithms relying on just human-defined rules and features are often difficult to extract the trail clearly when boundaries are ambiguous. In recent studies, the UAV automatically follows the forest trail by identifying the trail with the CNN [1, 2]. Most of the previous studies focus on the detection and following the trail only [1–9].
Data Loading...
