AiNavUT: an intelligent navigation system for unmanned tractor in airport

• Main Author: Zhou, Yichen
• Other Authors: Wang Dan Wei
• Format: Thesis-Master by Coursework
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Electrical and electronic engineering::Control and instrumentation::Robotics
• Online Access: https://hdl.handle.net/10356/164464
• Institution: Nanyang Technological University

With the growth in the number of airplanes at the airport, airport dispatch pressure will increase correspondingly. Airplane collisions frequently occur due to untimely communications and possible maloperations. To address this problem, AiNavUT, an intelligent navigation system for unmanned tractor has been proposed, where only a depth sensor is leveraged to perceive the environment and derive a low-price but robust autonomous navigation solution. AiNavUT consists of three main modules. Object detection: a multi-object detection and tracking is introduced to recognize the object. Localization: two methods based on poincloud clustering and monocular ranging are proposed to estimate depth in large baseline scenes. To improve the performance of monocular ranging, a semantic segmentation network is introduced. Navigation: based on the localization results, the costmap will be generated for the subsequent navigation.