公开(公告)号：US20240228077A1

公开(公告)日：2024-07-11

申请号：US18015321

申请日：2020-07-07

Applicant: INNER MONGOLIA UNIVERSITY OF TECHNOLOGY

Inventor： Yongsheng QI ,  Anyu CHEN ,  Yongting LI ,  Liqiang LIU ,  Songsong ZHANG

IPC: B64U50/39 ,  B60L53/51 ,  B60L53/80 ,  B64U70/97

CPC classification number: B64U50/39 ,  B60L53/51 ,  B60L53/80 ,  B64U70/97 ,  B60L2200/10

Abstract: The invention discloses an energy self-control base station for battery replacement based on solar power supply with independent UAV take-off and landing, which comprises a UAV take-off and landing device, an automatic battery replacement device and an energy supply device. The foldable solar panel structure is driven by a motor. The invention has simple structure and is convenient for automatic battery replacement. The UAV damage caused by UAV position deviation is effectively avoided when the platform descends. The invention adopts an electric mechanical claw to cooperate with a liftable battery compartment, so as to automatically replace UAV batteries. After the electric mechanical claw is aligned with the predetermined position, the batteries can be grasped and placed by lifting the battery compartment or the annular lifting platform, which simplifies the mechanical structure of the automatic battery replacement device.

公开(公告)号：US20230316699A1

公开(公告)日：2023-10-05

申请号：US18163918

申请日：2023-02-03

Applicant: INNER MONGOLIA UNIVERSITY OF TECHNOLOGY

Inventor： Yongsheng QI ,  Peiliang CHEN ,  Liqiang LIU ,  Yongting LI ,  Jianqiang SU

IPC: G06V10/26 ,  G06V10/80 ,  G06V10/40 ,  G06V10/82 ,  G06V10/776

CPC classification number: G06V10/26 ,  G06V10/40 ,  G06V10/82 ,  G06V10/776 ,  G06V10/806

Abstract: An image semantic segmentation algorithm and system based on multi-channel deep weighted aggregation where the image semantic segmentation algorithm is based on multi-channel deep weighted aggregation. The aggregation includes semantic features with definite class information in an image, transition semantic features between low-level semantic and high-level semantic, and semantic features of context logic relationship in an image are extracted by a low-level semantic channel, an auxiliary semantic channel and a high-level semantic channel, respectively. The aggregation further includes three different semantic features obtained in S1 are fused by weighted aggregation to obtain global semantic information of the image; S3, the semantic features output from respective semantic channels in S1 and the global semantic information in S2 are used to compute loss function for training.