XLXinyu Liu
Xinyu Liu
*This is an algorithmically generated author record based on publicly available information.
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Xinyu Liu has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Jie Qin1 shared publication
Kun Wang
College of Urban Transportation and Logistics, Shenzhen Technology University
1 shared publication- Luyao Jiang
College of Urban Transportation and Logistics, Shenzhen Technology University, Shenzhen 518118, China
1 shared publication - Peijiang Zhang
School of Information Engineering, Changan University, Xian, China
1 shared publication Tao Cheng
College of Urban Transportation and Logistics, Shenzhen Technology University
1 shared publication- Tao Cheng
College of Urban Transportation and Logistics, Shenzhen Technology University, Shenzhen 518118, China
1 shared publication
Latest Publications
- Open Access
- 15 August 2026
- Published by Elsevier in SSRN Electronic Journal
Gain--Cost Candidate Regulation for Efficient Multi-UAV Exploration under Sparse Sharing
Parallel sensing can shorten multi-UAV mapping missions, yet late-stage exploration remains costly when low-yield frontiers continue to trigger flight, reassignment, and communication. We formulate this problem as pre-allocation candidate regulation under sparse summary sharing and propose \method. Each frontier is represented by an inspectable record containing gain, viewpoint quality, travel cost, and peer-state information. Normalized gain evidence lowers the assignment cost within a fixed cap, and an auxiliary peer-connectivity term penalizes moves that increase the peer-distance burden. The allocator selects the eligible candidate with minimum adjusted cost; if none is available, the UAV retains its current target or receives no new assignment. We evaluate the method in three forest-like 3D simulation scenes using five matched runs per scene. Compared with the dense-communication reference (\dcr), final volume was $0.02\%$ lower, while completion time, flight distance, communication volume, and message count decreased by $12.72\%$, $12.01\%$, $37.54\%$, and $29.61\%$, respectively. Compared with the sparse-sharing baseline (\ssb), final volume was $0.08\%$ lower; completion time, flight distance, and message count decreased by $13.26\%$, $12.85\%$, and $7.32\%$, respectively. Communication volume decreased by $2.53\%$, but its confidence interval crossed zero. Module ablations indicate that gain evidence and reward capping both contribute to the observed regulation behavior. Removing the peer-connectivity term produced no significant paired change, so we treat it as an auxiliary regularizer.- 15 August 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessGain--Cost Candidate Regulation for Efficient Multi-UAV Exploration under Sparse Sharing
Parallel sensing can shorten multi-UAV mapping missions, yet late-stage exploration remains costly when low-yield frontiers continue to trigger flight, reassignment, and communication. We formulate this problem as pre-allocation candidate regulation under sparse summary sharing and propose \method. Each frontier is represented by an inspectable record containing gain, viewpoint quality, travel cost, and peer-state information. Normalized gain evidence lowers the assignment cost within a fixed cap, and an auxiliary peer-connectivity term penalizes moves that increase the peer-distance burden. The allocator selects the eligible candidate with minimum adjusted cost; if none is available, the UAV retains its current target or receives no new assignment. We evaluate the method in three forest-like 3D simulation scenes using five matched runs per scene. Compared with the dense-communication reference (\dcr), final volume was $0.02\%$ lower, while completion time, flight distance, communication volume, and message count decreased by $12.72\%$, $12.01\%$, $37.54\%$, and $29.61\%$, respectively. Compared with the sparse-sharing baseline (\ssb), final volume was $0.08\%$ lower; completion time, flight distance, and message count decreased by $13.26\%$, $12.85\%$, and $7.32\%$, respectively. Communication volume decreased by $2.53\%$, but its confidence interval crossed zero. Module ablations indicate that gain evidence and reward capping both contribute to the observed regulation behavior. Removing the peer-connectivity term produced no significant paired change, so we treat it as an auxiliary regularizer. - 4 November 2025
- Published by Taylor & Francis in International Journal of Human–Computer Interaction
- Vol. 42 (12) , 9606-9627
Cognitive Load in Novice UAV Pilots: A Preliminary fNIRS Investigation
- 4 November 2025
- Published by Taylor & Francis in International Journal of Human–Computer Interaction
- Vol. 42 (12) , 9606-9627
Cognitive Load in Novice UAV Pilots: A Preliminary fNIRS Investigation