pith. sign in

arxiv: 2604.09049 · v2 · submitted 2026-04-10 · 💻 cs.RO · cs.HC

TriDeliver: Cooperative Air-Ground Instant Delivery with UAVs, Couriers, and Crowdsourced Ground Vehicles

Junhui Gao , Yan Pan , Qianru Wang , Wenzhe Hou , Yiqin Deng , Liangliang Jiang , Yuguang Fang This is my paper

Pith reviewed 2026-05-10 17:57 UTC · model grok-4.3

classification 💻 cs.RO cs.HC
keywords instant deliverycooperative deliveryUAVscrowdsourced ground vehiclestransfer learningair-ground integrationdelivery scheduling
0
0 comments X

The pith

TriDeliver shows that transferring scheduling knowledge from couriers to UAVs and crowdsourced vehicles can cut instant delivery costs by more than 65 percent.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proposes a cooperative delivery system that combines human couriers, drones, and crowdsourced ground vehicles to handle instant deliveries more effectively than any group alone. It uses transfer learning to take scheduling patterns learned from couriers' past behavior and adapt them for the other agents. This matters because separate delivery methods each have limits in speed, cost, or availability, and better integration could meet growing demand without increasing expenses or worker burden. If the approach works as described, instant shipping becomes cheaper and quicker while reducing interference with the regular jobs of part-time drivers.

Core claim

TriDeliver introduces the first hierarchical cooperative framework for instant delivery that unites couriers, UAVs, and crowdsourced ground vehicles. A transfer learning algorithm extracts knowledge from couriers' delivery history and fine-tunes it for UAVs and GVs to improve parcel dispatching. On real one-month trajectory and delivery data, this yields a 65.8 percent lower delivery cost than existing UAV-courier methods, plus gains in speed, overall cost, and minimal disruption to crowdsourced vehicles' primary tasks.

What carries the argument

The transfer learning algorithm that extracts delivery scheduling knowledge from couriers' behavioral history and transfers it to UAVs and crowdsourced ground vehicles with fine-tuning for cooperative dispatching.

If this is right

  • Integrating couriers, UAVs, and crowdsourced ground vehicles outperforms prior two-agent cooperative methods in overall delivery cost.
  • Delivery times shorten and total costs decrease even when the transferred knowledge uses simple neural networks.
  • Crowdsourced ground vehicles see substantially lower interference with their original tasks.
  • The hierarchical framework enables efficient parcel dispatching across air and ground agents.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar transfer learning from experienced agents could extend to other multi-agent logistics tasks such as route planning for mixed vehicle fleets.
  • The system might scale to include additional delivery modes if more behavioral datasets become available for knowledge extraction.
  • Real-time updates to the transferred knowledge could further reduce delays in highly variable urban settings.

Load-bearing premise

The knowledge extracted from couriers' behavioral history will generalize effectively to UAVs and crowdsourced ground vehicles in dynamic real-world conditions.

What would settle it

A live deployment test that measures whether the 65.8 percent cost reduction and 17.7 percent time improvement hold when UAV flight restrictions, real-time courier availability, and actual crowdsourced vehicle participation rates replace historical data simulations.

Figures

Figures reproduced from arXiv: 2604.09049 by Junhui Gao, Liangliang Jiang, Qianru Wang, Wenzhe Hou, Yan Pan, Yiqin Deng, Yuguang Fang.

Figure 1
Figure 1. Figure 1: The Workflow of TriDeliver. scalable and cost-effective capacity boost without additional fixed investments. Furthermore, to overcome the ground traffic congestion that limits both couriers and GVs, UAVs are utilized to bypass jams and ensure the rapid delivery of urgent parcels. This hybrid approach minimizes operational costs while maximizing delivery efficiency and reliability across varying conditions.… view at source ↗
Figure 2
Figure 2. Figure 2: The Numbers of Delivery Or￾ders and Trips of GVs. However, the unstable sup￾ply of crowdsourced couri￾ers would lead to delayed delivery and degrade de￾livery performance in their work. Therefore, in this paper, we propose to re￾cruit dedicated couriers to collaborate with dedicated UAVs for parcel delivery [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: The Framework of TriDeliver [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Delivery Models of (a) UAVs and (b) couriers. [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Delivery Cases of GVs. (a) OD-pair Delivery; (b) Halfway Delivery;(c) Unoccupied Delivery. by taxis). These delivery cases by GVs are Origin-Destination (OD)-pair Delivery, Halfway Delivery, and Unoccupied De￾livery, respectively, shown in [PITH_FULL_IMAGE:figures/full_fig_p005_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Transfer and Fine-tuning of Decision Functions. [PITH_FULL_IMAGE:figures/full_fig_p006_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Impact of Delivery Demands on Delivery Performance [PITH_FULL_IMAGE:figures/full_fig_p009_7.png] view at source ↗
Figure 8
Figure 8. Figure 8: Impact of Ratio of Taxis Participating in Delivery on Delivery Performance [PITH_FULL_IMAGE:figures/full_fig_p009_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Impact of the Number of UAVs at Each Station on Delivery Performance [PITH_FULL_IMAGE:figures/full_fig_p009_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Impact of the Number of Couriers at Each Station on Delivery Performance [PITH_FULL_IMAGE:figures/full_fig_p009_10.png] view at source ↗
read the original abstract

Instant delivery, shipping items before critical deadlines, is essential in daily life. While multiple delivery agents, such as couriers, Unmanned Aerial Vehicles (UAVs), and crowdsourced agents, have been widely employed, each of them faces inherent limitations (e.g., low efficiency/labor shortages, flight control, and dynamic capabilities, respectively), preventing them from meeting the surging demands alone. This paper proposes TriDeliver, the first hierarchical cooperative framework, integrating human couriers, UAVs, and crowdsourced ground vehicles (GVs) for efficient instant delivery. To obtain the initial scheduling knowledge for GVs and UAVs as well as improve the cooperative delivery performance, we design a Transfer Learning (TL)-based algorithm to extract delivery knowledge from couriers' behavioral history and transfer their knowledge to UAVs and GVs with fine-tunings, which is then used to dispatch parcels for efficient delivery. Evaluated on one-month real-world trajectory and delivery datasets, it has been demonstrated that 1) by integrating couriers, UAVs, and crowdsourced GVs, TriDeliver reduces the delivery cost by $65.8\%$ versus state-of-the-art cooperative delivery by UAVs and couriers; 2) TriDeliver achieves further improvements in terms of delivery time ($-17.7\%$), delivery cost ($-9.8\%$), and impacts on original tasks of crowdsourced GVs ($-43.6\%$), even with the representation of the transferred knowledge by simple neural networks, respectively.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

3 major / 1 minor

Summary. The paper proposes TriDeliver, the first hierarchical cooperative framework integrating human couriers, UAVs, and crowdsourced ground vehicles (GVs) for instant delivery. It introduces a transfer learning (TL) algorithm in §3.2 that extracts scheduling knowledge from one-month courier behavioral history and transfers it to UAVs and GVs via fine-tuning, represented even by simple neural networks. Evaluated on real-world trajectory and delivery datasets, it claims a 65.8% delivery cost reduction versus state-of-the-art UAV+courier cooperative delivery, plus further gains of -17.7% in delivery time, -9.8% in cost, and -43.6% in impacts on GVs' original tasks.

Significance. If the empirical claims hold after addressing validation gaps, the work would be significant for multi-agent instant delivery systems by showing how TL can bridge domain differences across human, aerial, and ground agents in a hierarchical setup. It directly tackles efficiency limits of single-agent approaches and offers a practical path to cost and time reductions with minimal disruption to crowdsourced tasks, potentially informing real-world logistics platforms.

major comments (3)
  1. [§3.2] §3.2: The TL algorithm is load-bearing for the initial scheduling knowledge and all reported gains, yet the manuscript provides no ablation isolating the TL contribution (e.g., comparing the full TriDeliver against the hierarchical framework without TL or with random initialization), so the 65.8% cost reduction cannot be attributed specifically to transfer from courier data.
  2. [Evaluation section] Evaluation section (referenced in abstract): The concrete percentage improvements are presented without details on baseline implementations, statistical tests, error bars, data exclusion rules, or the exact training/validation split and fine-tuning procedure for the TL algorithm, which directly undermines confidence in the soundness of the headline claims.
  3. [§3.2] §3.2: The domain shift between courier 2D routing and UAV 3D kinematics/battery/airspace constraints (or GV traffic/capacity limits) is not addressed with any domain-adaptation metrics, sensitivity analysis, or perturbation tests; without these, the generalization assumption required for the -17.7% time and -43.6% GV-impact gains remains unverified.
minor comments (1)
  1. [Abstract] The abstract and evaluation description would benefit from explicit citation of the exact SOTA UAV+courier baseline method being compared against for the 65.8% figure.

Simulated Author's Rebuttal

3 responses · 0 unresolved

We thank the referee for the constructive and detailed feedback. The comments highlight important aspects of validating the transfer learning component and strengthening the evaluation rigor. We address each major comment below and will revise the manuscript to incorporate the suggested improvements.

read point-by-point responses

  1. Referee: [§3.2] The TL algorithm is load-bearing for the initial scheduling knowledge and all reported gains, yet the manuscript provides no ablation isolating the TL contribution (e.g., comparing the full TriDeliver against the hierarchical framework without TL or with random initialization), so the 65.8% cost reduction cannot be attributed specifically to transfer from courier data.

    Authors: We agree that an explicit ablation study isolating the transfer learning (TL) contribution would strengthen attribution of the reported gains. The current evaluation compares TriDeliver against state-of-the-art UAV+courier methods that lack courier-derived knowledge transfer. In the revision, we will add ablation experiments comparing the full framework against variants using random initialization and the hierarchical setup without TL, to directly quantify the TL component's impact on the 65.8% cost reduction. revision: yes

  2. Referee: [Evaluation section] The concrete percentage improvements are presented without details on baseline implementations, statistical tests, error bars, data exclusion rules, or the exact training/validation split and fine-tuning procedure for the TL algorithm, which directly undermines confidence in the soundness of the headline claims.

    Authors: We acknowledge that additional implementation and statistical details are necessary for full reproducibility and confidence in the results. The revised manuscript will expand the Evaluation section to include: detailed descriptions of all baseline implementations, statistical significance tests (e.g., paired t-tests), error bars on performance figures, explicit data exclusion and preprocessing rules, the precise training/validation splits, and the full fine-tuning procedure with hyperparameters for the TL algorithm. revision: yes

  3. Referee: [§3.2] The domain shift between courier 2D routing and UAV 3D kinematics/battery/airspace constraints (or GV traffic/capacity limits) is not addressed with any domain-adaptation metrics, sensitivity analysis, or perturbation tests; without these, the generalization assumption required for the -17.7% time and -43.6% GV-impact gains remains unverified.

    Authors: The TL algorithm employs fine-tuning specifically to adapt courier-derived scheduling knowledge to the distinct constraints of UAVs (3D kinematics, battery, airspace) and GVs (traffic, capacity). The real-world trajectory evaluations demonstrate effective transfer. To further substantiate generalization, the revision will add domain-adaptation metrics (e.g., pre- vs. post-fine-tuning performance deltas) and sensitivity analyses under perturbations to UAV/GV constraints. revision: yes

Circularity Check

0 steps flagged

No circularity: performance claims rest on external real-world dataset evaluation

full rationale

The paper presents an empirical system evaluated on one-month real-world trajectory and delivery datasets. The TL algorithm extracts scheduling knowledge from courier behavioral history and transfers it via fine-tuning to UAVs and GVs; reported gains (65.8% cost reduction vs. SOTA, plus time/cost/GV-impact improvements) are measured outcomes on held-out data rather than quantities defined in terms of the model's own fitted parameters or self-referential equations. No self-definitional steps, fitted-input-called-prediction patterns, load-bearing self-citations, uniqueness theorems, or ansatzes smuggled via citation appear in the abstract or described derivation. The central claims remain independent of the paper's own outputs and are falsifiable against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 2 axioms · 0 invented entities

Only the abstract is available, so the ledger captures the main implicit assumptions required for the central claims to hold. No explicit free parameters or invented entities are named. The transfer learning step and crowdsourced vehicle availability rest on domain assumptions rather than derived results.

axioms (2)
  • domain assumption Crowdsourced ground vehicles can be dispatched for deliveries with only modest impact on their original tasks.
    The reported -43.6% impact reduction presupposes reliable availability and cooperation of GVs.
  • domain assumption Behavioral patterns from courier delivery history transfer usefully to UAVs and GVs after fine-tuning.
    This is the core premise of the TL-based algorithm that enables the claimed performance gains.

pith-pipeline@v0.9.0 · 5603 in / 1614 out tokens · 55456 ms · 2026-05-10T17:57:22.100420+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Collaborative Air-Ground Sensing, Communication, Computing, Storage, and Intelligence for Low-Altitude Economy

    cs.NI 2026-05 unverdicted novelty 4.0

    The paper proposes a task-driven framework connecting LAE scenarios, air-ground architecture, and SCCSI co-optimization toolboxes for collaborative low-altitude systems.

Reference graph

Works this paper leans on

66 extracted references · 66 canonical work pages · cited by 1 Pith paper · 1 internal anchor

  1. [1]

    Faircod: A fairness-aware concurrent dispatch system for large-scale instant delivery services,

    L. Jiang, S. Wang, B. Guo, H. Wang, D. Zhang, and G. Wang, “Faircod: A fairness-aware concurrent dispatch system for large-scale instant delivery services,” inProceedings of the 29th ACM SIGKDD Conference on knowledge discovery and data mining, 2023, pp. 4229–4238

    work page 2023

  2. [2]

    Emergency medical services,

    O. C. Kobusingye, A. A. Hyder, D. Bishai, M. Joshipura, E. R. Hicks, and C. Mock, “Emergency medical services,”Disease Control Priorities in Developing Countries. 2nd edition, 2006

    work page 2006

  3. [3]

    (2023) 41+ global online food delivery statistics (2023)

    Otter. (2023) 41+ global online food delivery statistics (2023). [Online]. Available: https://www.tryotter.com/en-gb/blog/ industry/online-food-delivery-statistics

    work page 2023

  4. [4]

    P. C. Group. Instant delivery market analysis. [Online]. Available: https://pmarketresearch.com/instant-delivery-market-analysis/

    work page

  5. [5]

    D. Curry. Food delivery app revenue and usage statistics (2025). [Online]. Available: https://www.businessofapps.com/data/ food-delivery-app-market/

    work page 2025

  6. [6]

    Y . Wang. Jd’s self-developed logistics drone takes off in sichuan. [Online]. Available: https://jdcorporateblog.com/ jds-self-developed-logistics-drone-takes-off-in-sichuan/

    work page

  7. [7]

    Online crowdsourced truck delivery using historical information,

    H. Zhang, K. Luo, Y . Xu, Y . Xu, and W. Tong, “Online crowdsourced truck delivery using historical information,”European Journal of Operational Research, vol. 301, no. 2, pp. 486–501, 2022. [Online]. Available: https://www.sciencedirect.com/science/article/pii/ S0377221721008869

    work page 2022

  8. [8]

    M. T. Review. (2023) Drone food delivery is now part of daily life in shenzhen. [Online]. Available: https://www.technologyreview.com/ 2023/05/23/1073500/drone-food-delivery-shenzhen-meituan/

    work page 2023

  9. [9]

    A. Flex. How delivering pakages with amazon flex works. [Online]. Available: https://flex.amazon.com/lets-drive

    work page

  10. [10]

    Order with amazon prime drone delivery

    Amazon. Order with amazon prime drone delivery. [Online]. Avail- able: https://www.amazon.com/gp/help/customer/display.html?nodeId= TeqJgaugxFtL4Lj7hy

    work page

  11. [11]

    Aerial picking and delivery of magnetic objects with mavs,

    A. Gawel, M. Kamel, T. Novkovic, J. Widauer, D. Schindler, B. P. V on Altishofen, R. Siegwart, and J. Nieto, “Aerial picking and delivery of magnetic objects with mavs,” in2017 IEEE international conference on robotics and automation (ICRA). IEEE, 2017, pp. 5746–5752

    work page 2017

  12. [12]

    Review of navigation methods for uav-based parcel delivery,

    D. Dissanayaka, T. R. Wanasinghe, O. De Silva, A. Jayasiri, and G. K. Mann, “Review of navigation methods for uav-based parcel delivery,” IEEE Transactions on Automation Science and Engineering, 2023

    work page 2023

  13. [13]

    Efficient schedule of energy-constrained uav using crowd- sourced buses in last-mile parcel delivery,

    Y . Pan, S. Li, Q. Chen, N. Zhang, T. Cheng, Z. Li, B. Guo, Q. Han, and T. Zhu, “Efficient schedule of energy-constrained uav using crowd- sourced buses in last-mile parcel delivery,”Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, vol. 5, no. 1, pp. 1–23, 2021

    work page 2021

  14. [14]

    Nationwide deployment and operation of a virtual arrival detection system in the wild,

    Y . Ding, Y . Yang, W. Jiang, Y . Liu, T. He, and D. Zhang, “Nationwide deployment and operation of a virtual arrival detection system in the wild,” inProceedings of the 2021 ACM SIGCOMM 2021 Conference, 2021, pp. 705–717

    work page 2021

  15. [15]

    Recommending-and-grabbing: A crowdsourcing-based order allocation pattern for on-demand food delivery,

    X. Wang, L. Wang, S. Wang, J. Pan, H. Ren, and J. Zheng, “Recommending-and-grabbing: A crowdsourcing-based order allocation pattern for on-demand food delivery,”IEEE Transactions on Intelligent Transportation Systems, vol. 24, no. 1, pp. 838–853, 2022

    work page 2022

  16. [16]

    A city-wide crowdsourcing delivery system with reinforcement learning,

    Y . Ding, B. Guo, L. Zheng, M. Lu, D. Zhang, S. Wang, S. H. Son, and T. He, “A city-wide crowdsourcing delivery system with reinforcement learning,”Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, vol. 5, no. 3, pp. 1–22, 2021

    work page 2021

  17. [17]

    Transfloor: Transparent floor localization for crowdsourcing instant delivery,

    Z. Xie, H. Luo, X. Zhang, H. Xiong, F. Zhao, Z. Li, Q. Ye, B. Rong, and J. Gao, “Transfloor: Transparent floor localization for crowdsourcing instant delivery,”Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, vol. 6, no. 4, pp. 1–30, 2023

    work page 2023

  18. [18]

    Pioneering cooperative air-ground instant delivery using uavs and crowdsourced couriers,

    Y . Pan, J. Gao, J. Duan, J. Shi, B. Guo, Y . Liang, and Y . Hu, “Pioneering cooperative air-ground instant delivery using uavs and crowdsourced couriers,”Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies, vol. 8, no. 4, pp. 1–26, 2024

    work page 2024

  19. [19]

    Joint trajectory and pick- up design for uav-assisted item delivery under no-fly zone constraints,

    W. Wen, K. Luo, L. Liu, Y . Zhang, and Y . Jia, “Joint trajectory and pick- up design for uav-assisted item delivery under no-fly zone constraints,” IEEE Transactions on Vehicular Technology, vol. 72, no. 2, pp. 2587– 2592, 2022

    work page 2022

  20. [20]

    Multiuser miso uav communications in uncertain environments with no-fly zones: Robust trajectory and resource allocation design,

    D. Xu, Y . Sun, D. W. K. Ng, and R. Schober, “Multiuser miso uav communications in uncertain environments with no-fly zones: Robust trajectory and resource allocation design,”IEEE Transactions on Com- munications, vol. 68, no. 5, pp. 3153–3172, 2020

    work page 2020

  21. [21]

    There are millions of jobs, but a short- age of workers: Economists explain why that’s wor- rying

    CNBC. There are millions of jobs, but a short- age of workers: Economists explain why that’s wor- rying. [Online]. Available: https://www.cnbc.com/2021/10/20/ global-shortage-of-workers-whats-going-on-experts-explain.html

    work page 2021

  22. [23]

    (2022) Why instant delivery is getting slower?(in chinese)

    Shobserver. (2022) Why instant delivery is getting slower?(in chinese). [Online]. Available: https://new.qq.com/rain/a/20221220A08AAN00

    work page arXiv 2022

  23. [24]

    An exact solution method for the capacitated item-sharing and crowdshipping problem,

    M. Behrend, F. Meisel, K. Fagerholt, and H. Andersson, “An exact solution method for the capacitated item-sharing and crowdshipping problem,”European Journal of Operational Research, vol. 279, no. 2, pp. 589–604, 2019

    work page 2019

  24. [25]

    Cooperative air-ground instant delivery by uavs and crowdsourced 12 taxis,

    J. Gao, Q. Wang, X. Zhang, J. Shi, X. Zhao, Q. Han, and Y . Pan, “Cooperative air-ground instant delivery by uavs and crowdsourced 12 taxis,” in2024 IEEE 40th International Conference on Data Engineering (ICDE), 2024, pp. 4153–4166

    work page 2024

  25. [26]

    Drone routing in a time- dependent network: Toward low-cost and large-range parcel delivery,

    H. Huang, A. V . Savkin, and C. Huang, “Drone routing in a time- dependent network: Toward low-cost and large-range parcel delivery,” IEEE Transactions on Industrial Informatics, vol. 17, no. 2, pp. 1526– 1534, 2020

    work page 2020

  26. [27]

    Crowdsourcing last-mile deliveries,

    S. Fatehi and M. R. Wagner, “Crowdsourcing last-mile deliveries,” Manufacturing & Service Operations Management, vol. 24, no. 2, pp. 791–809, 2022

    work page 2022

  27. [28]

    Cooperative air-ground instant delivery by uavs and crowdsourced taxis: Joint uav station deployment and delivery scheduling,

    J. Gao, Q. Wang, X. Zhang, J. Shi, X. Zhao, Y . Liang, B. Guo, Q. Han, and Y . Pan, “Cooperative air-ground instant delivery by uavs and crowdsourced taxis: Joint uav station deployment and delivery scheduling,”IEEE Transactions on Mobile Computing, pp. 1–17, 2025

    work page 2025

  28. [29]

    Velp: vehicle loading plan learning from human behavior in nationwide logistics system,

    S. Duan, F. Lyu, X. Zhu, Y . Ding, H. Wang, D. Zhang, X. Liu, Y . Zhang, and J. Ren, “Velp: vehicle loading plan learning from human behavior in nationwide logistics system,”Proceedings of the VLDB Endowment, vol. 17, no. 2, pp. 241–249, 2023

    work page 2023

  29. [30]

    Efficient package delivery task assignment for truck and high capacity drone,

    X. Bai, Y . Ye, B. Zhang, and S. S. Ge, “Efficient package delivery task assignment for truck and high capacity drone,”IEEE Transactions on Intelligent Transportation Systems, 2023

    work page 2023

  30. [31]

    First trial of drone parcel delivery con- ducted

    ePlane AI. First trial of drone parcel delivery con- ducted. [Online]. Available: https://www.eplaneai.com/nl/news/ first-trial-of-drone-parcel-delivery-conducted

    work page

  31. [32]

    Potentialities of drones and ground autonomous delivery devices for last- mile logistics,

    C. Lemardel ´e, M. Estrada, L. Pag `es, and M. Bachofner, “Potentialities of drones and ground autonomous delivery devices for last- mile logistics,”Transportation Research Part E: Logistics and Transportation Review, vol. 149, p. 102325, 2021. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S1366554521000995

    work page 2021

  32. [33]

    UA V-enabled computing power networks: Design and performance analysis under energy constraints,

    Y . Deng, Z. Fang, S. Hu, Y . Ma, X. Guo, H. Zhang, and Y . Fang, “UA V-enabled computing power networks: Design and performance analysis under energy constraints,”IEEE Transactions on Mobile Computing, p. 1–17, 2026. [Online]. Available: http: //dx.doi.org/10.1109/TMC.2026.3655118

    work page doi:10.1109/tmc.2026.3655118 2026

  33. [34]

    Image-based visual servoing of unmanned aerial manipulators for tracking and grasping a moving target,

    Y . Chen, Y . Wu, Z. Zhang, Z. Miao, H. Zhong, H. Zhang, and Y . Wang, “Image-based visual servoing of unmanned aerial manipulators for tracking and grasping a moving target,”IEEE Transactions on Industrial Informatics, 2022

    work page 2022

  34. [35]

    Aerial picking and delivery of magnetic objects with mavs,

    A. Gawel, M. Kamel, T. Novkovic, J. Widauer, D. Schindler, B. P. von Altishofen, R. Siegwart, and J. Nieto, “Aerial picking and delivery of magnetic objects with mavs,” in2017 IEEE International Conference on Robotics and Automation (ICRA), 2017, pp. 5746–5752

    work page 2017

  35. [36]

    Uav trajectory planning with probabilistic geo-fence via iterative chance- constrained optimization,

    B. Du, J. Chen, D. Sun, S. G. Manyam, and D. W. Casbeer, “Uav trajectory planning with probabilistic geo-fence via iterative chance- constrained optimization,”IEEE Transactions on Intelligent Transporta- tion Systems, vol. 23, no. 6, pp. 5859–5870, 2021

    work page 2021

  36. [37]

    Vehicle routing problems for drone delivery,

    K. Dorling, J. Heinrichs, G. G. Messier, and S. Magierowski, “Vehicle routing problems for drone delivery,”IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 47, no. 1, pp. 70–85, 2017

    work page 2017

  37. [38]

    A method of optimized deployment of charging stations for drone delivery,

    H. Huang and A. V . Savkin, “A method of optimized deployment of charging stations for drone delivery,”IEEE Transactions on Transporta- tion Electrification, vol. 6, no. 2, pp. 510–518, 2020

    work page 2020

  38. [39]

    A 500-w wireless charging system with lightweight pick-up for unmanned aerial vehicles,

    C. Cai, S. Wu, L. Jiang, Z. Zhang, and S. Yang, “A 500-w wireless charging system with lightweight pick-up for unmanned aerial vehicles,” IEEE Transactions on Power Electronics, 2020

    work page 2020

  39. [40]

    A model and algorithm for the courier delivery problem with uncertainty,

    I. Sungur, Y . Ren, F. Ord ´o˜nez, M. Dessouky, and H. Zhong, “A model and algorithm for the courier delivery problem with uncertainty,” Transportation science, vol. 44, no. 2, pp. 193–205, 2010

    work page 2010

  40. [41]

    Courier dispatch in on-demand delivery,

    M. Chen and M. Hu, “Courier dispatch in on-demand delivery,”Man- agement Science, vol. 70, no. 6, pp. 3789–3807, 2024

    work page 2024

  41. [42]

    Dynamic courier capacity acquisition in rapid delivery systems: A deep q-learning approach,

    R. Auad, A. Erera, and M. Savelsbergh, “Dynamic courier capacity acquisition in rapid delivery systems: A deep q-learning approach,” Transportation Science, vol. 58, no. 1, pp. 67–93, 2024

    work page 2024

  42. [43]

    Courier routing and assignment for food delivery service using reinforcement learning,

    A. Bozanta, M. Cevik, C. Kavaklioglu, E. M. Kavuk, A. Tosun, S. B. Sonuc, A. Duranel, and A. Basar, “Courier routing and assignment for food delivery service using reinforcement learning,”Computers And Industrial Engineering, vol. 164, p. 107871, 2022

    work page 2022

  43. [44]

    A predict- then-optimize couriers allocation framework for emergency last-mile logistics,

    K. Xia, L. Lin, S. Wang, H. Wang, D. Zhang, and T. He, “A predict- then-optimize couriers allocation framework for emergency last-mile logistics,” inProceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, ser. KDD ’23. New York, NY , USA: Association for Computing Machinery, 2023, p. 5237–5248

    work page 2023

  44. [45]

    Come: Learning to coordinate crowdsourcing and regular couriers for offline delivery during online mega sale days,

    G. Zhu, D. Zhao, Y . Wang, H. Wang, D. Zhang, and H. Ma, “Come: Learning to coordinate crowdsourcing and regular couriers for offline delivery during online mega sale days,” in2023 IEEE 39th International Conference on Data Engineering (ICDE), 2023, pp. 3126–3139

    work page 2023

  45. [46]

    Rede: Exploring relay transportation for efficient last-mile delivery,

    W. Lyu, H. Wang, Z. Hong, G. Wang, Y . Yang, Y . Liu, and D. Zhang, “Rede: Exploring relay transportation for efficient last-mile delivery,” in 2023 IEEE 39th International Conference on Data Engineering (ICDE), 2023, pp. 3003–3016

    work page 2023

  46. [47]

    Foodnet: Toward an optimized food delivery network based on spatial crowdsourcing,

    Y . Liu, B. Guo, C. Chen, H. Du, Z. Yu, D. Zhang, and H. Ma, “Foodnet: Toward an optimized food delivery network based on spatial crowdsourcing,”IEEE Transactions on Mobile Computing, vol. 18, no. 6, pp. 1288–1301, 2018

    work page 2018

  47. [48]

    Fair task allocation in crowdsourced delivery,

    F. Basık, B. Gedik, H. Ferhatosmano ˘glu, and K.-L. Wu, “Fair task allocation in crowdsourced delivery,”IEEE Transactions on Services Computing, vol. 14, no. 4, pp. 1040–1053, 2018

    work page 2018

  48. [49]

    Crowdsourcing the last mile delivery of online orders by exploiting the social networks of retail store customers,

    A. Devari, A. G. Nikolaev, and Q. He, “Crowdsourcing the last mile delivery of online orders by exploiting the social networks of retail store customers,”Transportation Research Part E: Logistics and Transporta- tion Review, vol. 105, pp. 105–122, 2017

    work page 2017

  49. [50]

    Online delivery route recommendation in spatial crowdsourcing,

    D. Sun, K. Xu, H. Cheng, Y . Zhang, T. Song, R. Liu, and Y . Xu, “Online delivery route recommendation in spatial crowdsourcing,”World Wide Web, vol. 22, pp. 2083–2104, 2019

    work page 2083

  50. [51]

    Measuring maximum urban capacity of taxi-based logistics,

    Y . Chen, D. Guo, M. Xu, G. Tang, and G. Cheng, “Measuring maximum urban capacity of taxi-based logistics,”IEEE Transactions on Intelligent Transportation Systems, vol. 22, no. 10, pp. 6449–6459, 2020

    work page 2020

  51. [52]

    Effi- cient large-scale multi-drone delivery using transit networks,

    S. Choudhury, K. Solovey, M. J. Kochenderfer, and M. Pavone, “Effi- cient large-scale multi-drone delivery using transit networks,”Journal of Artificial Intelligence Research, vol. 70, pp. 757–788, 2021

    work page 2021

  52. [53]

    Drone routing in a time- dependent network: Toward low-cost and large-range parcel delivery,

    H. Huang, A. V . Savkin, and C. Huang, “Drone routing in a time- dependent network: Toward low-cost and large-range parcel delivery,” IEEE Transactions on Industrial Informatics, vol. 17, no. 2, pp. 1526– 1534, 2021

    work page 2021

  53. [54]

    From conception to retirement: A lifetime story of a 3-year-old wireless beacon system in the wild,

    Y . Ding, L. Liu, Y . Yang, Y . Liu, D. Zhang, and T. He, “From conception to retirement: A lifetime story of a 3-year-old wireless beacon system in the wild,”IEEE/ACM Transactions on Networking, vol. 30, no. 1, pp. 47–61, 2022

    work page 2022

  54. [55]

    Transloc: Transparent indoor localization with uncertain human participation for instant delivery,

    Y . Yang, Y . Ding, D. Yuan, G. Wang, X. Xie, Y . Liu, T. He, and D. Zhang, “Transloc: Transparent indoor localization with uncertain human participation for instant delivery,” inProceedings of the 26th Annual International Conference on Mobile Computing and Networking, ser. MobiCom ’20. New York, NY , USA: Association for Computing Machinery, 2020

    work page 2020

  55. [56]

    Stl: Online detection of taxi trajectory anomaly based on spatial-temporal laws

    cbdog94. Stl: Online detection of taxi trajectory anomaly based on spatial-temporal laws. [Online]. Available: https://github.com/cbdog94/ STL

    work page

  56. [57]

    Extending delivery range and decelerating battery aging of logistics uavs using public buses,

    Y . Pan, Q. Chen, N. Zhang, Z. Li, T. Zhu, and Q. Han, “Extending delivery range and decelerating battery aging of logistics uavs using public buses,”IEEE Transactions on Mobile Computing, vol. 22, no. 9, pp. 5280–5295, 2023

    work page 2023

  57. [58]

    S. B. of Justice. Management regulation for civil drones in shenzheng (in chinese). [Online]. Available: https://sf.sz.gov.cn/xxgk/xxgkml/gsgg/ content/post 9276739.html

    work page

  58. [59]

    Drive less but finish more: Food delivery based on multi-level workers in spatial crowdsourcing,

    X. Xu, A. Liu, G. Liu, Z. Li, and L. Zhao, “Drive less but finish more: Food delivery based on multi-level workers in spatial crowdsourcing,” in Proceedings of the 31st ACM International Conference on Information & Knowledge Management, 2022, pp. 2331–2340

    work page 2022

  59. [60]

    crowddeliver: Planning city-wide package delivery paths leveraging the crowd of taxis,

    C. Chen, D. Zhang, X. Ma, B. Guo, L. Wang, Y . Wang, and E. Sha, “crowddeliver: Planning city-wide package delivery paths leveraging the crowd of taxis,”IEEE Transactions on Intelligent Transportation Systems, vol. 18, no. 6, pp. 1478–1496, 2017

    work page 2017

  60. [61]

    Reusing delivery drones for urban crowdsensing,

    C. Xiang, Y . Zhou, H. Dai, Y . Qu, S. He, C. Chen, and P. Yang, “Reusing delivery drones for urban crowdsensing,”IEEE Transactions on Mobile Computing, 2021

    work page 2021

  61. [62]

    L. Wang. Meituan adjusts takeout delivery commission draw: Billing by distance, price, time slot. [Online]. Available: https: //m.thepaper.cn/newsDetail forward 12619693

    work page

  62. [63]

    The share- a-ride problem: People and parcels sharing taxis,

    B. Li, D. Krushinsky, H. A. Reijers, and T. Van Woensel, “The share- a-ride problem: People and parcels sharing taxis,”European Journal of Operational Research, vol. 238, no. 1, pp. 31–40, 2014

    work page 2014

  63. [64]

    S. M. T. Commission. Taxi fare structure and tariffs of shanghai. [Online]. Available: https://jtw.sh.gov.cn/czqcyj/20180605/0010-10460. html

    work page arXiv

  64. [65]

    Adam: A Method for Stochastic Optimization

    D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” arXiv preprint arXiv:1412.6980, 2014

    work page internal anchor Pith review Pith/arXiv arXiv 2014

  65. [66]

    An efficient approximation for the generalized assignment problem,

    R. Cohen, L. Katzir, and D. Raz, “An efficient approximation for the generalized assignment problem,”Information Processing Letters, vol. 100, no. 4, pp. 162–166, 2006

    work page 2006

  66. [67]

    Characteristics of dji mavic 3 pro

    DJI. Characteristics of dji mavic 3 pro. [Online]. Available: https://www.dji.com/cn/mavic-3-pro/specs

    work page