Study on Aircraft Wing Collision Avoidance through Vision-Based Trajectory Prediction Civil Aviation University of China

Format:

Book

Conference/Event

Author/Creator:

Zhu, Hengjia, author.

Contributor:

JiYuan, Liu

Xu, Yitong

Xu, ZiShuo

Zhang, Wei

Conference Name:

WCX SAE World Congress Experience (2024-04-16 : Detroit, Michigan, United States)

Language:

English

Physical Description:

1 online resource cm

Place of Publication:

Warrendale, PA SAE International 2024

Summary:

When the aircraft towing operations are carried out in narrow areas such as the hangars or parking aprons, it has a high safety risk for aircraft that the wingtips may collide with the surrounding aircraft or the airport facility. A real-time trajectory prediction method for the towbarless aircraft taxiing system (TLATS) is proposed to evaluate the collision risk based on image recognition. The Yolov7 module is utilized to detect objects and extract the corresponding features. By obtaining information about the configuration of the airplane wing and obstacles in a narrow region, a Long Short-Term Memory (LSTM) encoder-decoder model is utilized to predict future motion trends. In addition, a video dataset containing the motions of various airplane wings in real traction scenarios is constructed for training and testing. Compared with the conventional methods, the proposed method combines image recognition and trajectory prediction methods to describe the relative positional relationship between the wings and obstacles, which enhances the accuracy of aircraft wing collision prediction during aircraft towing operations

Notes:

Vendor supplied data

Publisher Number:

2024-01-2310

Access Restriction:

Restricted for use by site license