Pith. sign in

REVIEW 5 major objections 7 minor 36 references

FM-Planner: Foundation Model Guided Path Planning for Autonomous Drone Navigation

T0 review · 5 major / 7 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Fine-tuned Llama matches A* on drone path planning

desk verdict A useful empirical benchmark and a real flight demo, but the paper's central 'robust obstacle awareness' claim outruns the evidence—especially the unvalidated YOLO+depth perception chain. read the letter →

arxiv 2505.20783 v1 pith:R3OZIA66 submitted 2025-05-27 cs.RO cs.AI

Jiaping Xiao , Cheng Wen Tsao , Yuhang Zhang , Mir Feroskhan This is my paper · ORCID
classification cs.ROcs.AI
keywords foundationmodelpathplanningautonomousdronelargelanguagevision-languageLoRAfine-tuningwaypointgenerationobstacleavoidance
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper sets out to show that foundation models can do global path planning for drones, not just local reactive control. It benchmarks eight LLMs and five VLMs on simulated obstacle-avoidance tasks and finds that pure LLMs operating on textual coordinates outperform VLMs that must reason from map images. Its central result is an LLM-Vision planner: a LoRA-fine-tuned Llama-3.1-8B-Instruct fed by a YOLOv8 vision encoder, which reaches 100 percent success in the two- and three-obstacle simulation scenarios and completes all three real-world flights with about 9.5 seconds of average reasoning time. If the result holds, it means a modest, fine-tuned language model paired with an off-the-shelf object detector can replace classical planners such as A* in known static indoor environments.

What carries the argument

The load-bearing object is the tokenized prompt $T_{\mathrm{input}} = \langle \text{Prompt}, P_s, P_g, \{P_o^n\}_{n=1}^N\rangle$, which serializes start, goal, and obstacle coordinates into natural language for the LLM. A YOLOv8 vision encoder turns the drone's RGB-D stream into class-labeled 2D bounding boxes and estimated 3D positions, which are tokenized into that prompt. The LLM is fine-tuned with Low-Rank Adaptation (LoRA), keeping the base weights frozen and learning only low-rank updates $\Delta W = BA$, which adds $O(dr + rk)$ parameters instead of $O(dk)$. A Euclidean-interpolation module inserts waypoints every 0.5 m and keeps a 1.2 m safety margin around obstacles; in the physical system the margin is 1-1.2 m. This pipeline converts sparse language-generated waypoints into smooth, flight-ready position setpoints.

What would settle it

Measure the actual 3D position error of the YOLOv8 detection pipeline against a ground-truth motion-capture of obstacle locations in the same arena; if the error is comparable to or larger than the 1-1.2 m safety margin, then the planner's collision-free guarantee fails. A simpler version: perturb the obstacle coordinates fed to the LLM by 0.5-1.0 m in simulation and record whether success rate collapses.

Watch

Extended reading notes

Core claim

The paper claims that a fine-tuned LLM integrated with a vision encoder demonstrates robust spatial reasoning and real-time obstacle awareness, making it well-suited for practical drone global planning tasks. In the simulation benchmarks, the fine-tuned Llama-3.1-8B-Instruct matches the optimal A* planner on both reliability and efficiency, recording a 100 percent success rate and efficiency-success scores of 1.250 and 1.111 on the two- and three-obstacle scenarios; it is the only neural planner to do so. In physical experiments, the LLM-Vision planner guided a QAV250 racing drone through all three test tasks with different start positions and obstacle configurations, with a mean reasoning time near 9.5 seconds. The authors interpret this as evidence that textual spatial inputs plus separately perceived obstacle locations generalize better than pure VLM visual reasoning.

Load-bearing premise

The whole safety argument rests on the unvalidated premise that the YOLOv8-plus-depth obstacle positions are accurate enough that the 1-1.2 m clearance margins truly keep the drone collision-free.

Editorial extensions

If this is right

  • A small LoRA-fine-tuned LLM can serve as a global path planner on edge hardware, generating collision-free waypoints without explicit dynamics models.
  • Classical planners like A* remain the optimality reference, but the LLM route adds the ability to absorb natural-language mission constraints into the same planning step.
  • The planner's reliability degrades sharply at control rates above 50 Hz, so the practical speed envelope is bounded by re-planning latency.
  • VLMs alone produced non-optimal or colliding paths in simulation, indicating that visual grounding must be supplied separately to the LLM rather than inside a single VLM.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The paper's benchmark is restricted to two static obstacle configurations; a natural stress test is whether the fine-tuned model generalizes to arbitrary obstacle counts and positions without re-tuning.
  • The absence of quantified perception error leaves open whether the same pipeline would be safe in tighter spaces where the 1.2 m margin is unavailable.
  • Because the synthetic fine-tuning data were generated with RRT, the policy may inherit sampling artifacts rather than learning true geometric reasoning; this would show up in corridors narrower than the training distribution.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 7 minor

Summary. This paper proposes FM-Planner, a foundation-model-based global path planning framework for drones. The authors benchmark eight LLMs and five VLMs in a Gazebo simulation with static cubic obstacles, using A*, RRT, and Q-learning as baselines and reporting success rate (SR), average completion time (ACT), path length (PL), and an efficiency-success score (ESS). Based on the benchmark, they fine-tune Llama-3.1-8B-Instruct with LoRA on 5,000 synthetic waypoint sequences generated by RRT and integrate it with a YOLOv8 vision encoder to estimate obstacle positions from RGB-D images. They validate the resulting LLM-Vision planner in three indoor flight tasks with a QAV250 drone and report that all three tasks succeeded with an average reasoning time of about 9.5 s per query. The headline claims are that Llama-3.1-8B-Instruct matches the A* planner on reliability and efficiency in simulation, and that the fine-tuned LLM-Vision system demonstrates robust spatial reasoning and real-time obstacle awareness.

Significance. The paper addresses a timely question—whether foundation models can perform global path planning for drones—and the base-LLM benchmark is a useful contribution: it compares eight models against classical planners in a standardized setting, and the independent comparison with external baselines makes the headline simulation result non-circular. The open project site and detailed hyperparameter table also support reproducibility. However, the significance of the physical-validation claim is currently limited. The simulation results are reported as point estimates without variance or trial counts, the VLM evaluation is qualitative, and the real-world experiments provide no error analysis for the perception pipeline or comparison with baselines. As presented, the evidence does not support the conclusion that the system has robust, real-time obstacle awareness.

major comments (5)
  1. [III-C, IV-G, Table V] The central safety claim rests on an unvalidated perception pipeline. Section III-C states that YOLOv8 plus depth information is used to 'estimate the 3D position' of obstacles, and Section IV-G reports that obstacle information was extracted from RGB and depth images during flight. However, the paper provides no detection accuracy, depth-error analysis, calibration check, or failure cases for this pipeline. Table V reports ground-truth obstacle coordinates, not the estimated coordinates actually fed to the LLM. With the only buffers being the 1.2 m margin in Section III-A and the 1 m clearance in Section IV-F, a misdetection or depth bias larger than these margins would produce unsafe waypoints while the run is still counted as a success. The three single-shot physical tasks therefore support only the narrow claim that the drone succeeded in three particular configurations, not the paper's conclusion of robust obstacle awareness.
  2. [IV-E] The VLM benchmark is not quantified. Section IV-E2 describes the VLM-generated trajectories only qualitatively (e.g., 'GPT-4o and Claude produced relatively coherent and feasible paths'), and Fig. 5 shows paths without any numerical metrics. Since the abstract and introduction claim a 'comprehensive benchmarking study' of LLM and VLM approaches, the VLM portion needs at least the same SR/PL/ESS metrics and trial counts as the LLM portion to support the comparison and the claim that VLMs lack spatial reasoning.
  3. [IV-D, Tables I-III] The simulation benchmark reports single values without variance, number of trials, or random seeds. The text mentions '10 trials' only in Section IV-D1, and Table II shows one 0/10 entry, but the SR differences that drive the paper's conclusions (e.g., 100% vs. 90% for Llama vs. Qwen-7B in S1) are not statistically distinguishable at n=10. Please report per-trial data, confidence intervals, and seeds, or increase the number of trials, for all reported SR/ACT/PL/ESS values.
  4. [IV-F, IV-G, V] The 'real-time obstacle awareness' claim is not supported by the reported 9.5 s reasoning time and the described architecture. Section IV-G states that the LLM is executed at each planning step, but it does not specify how often the YOLOv8 perception updates the obstacle list or whether the global path is recomputed during flight. If the waypoint sequence is generated once before takeoff, then the system is not providing real-time obstacle awareness in any feedback sense; it is a one-shot global planner with offline perception. The conclusion should be qualified accordingly, or the paper should report the update rate and show that the planner reacts to perception changes during flight.
  5. [IV-F] The fine-tuning evaluation is incomplete. The LLM-Vision model is trained on 5,000 synthetic scenarios whose waypoints were computed by RRT (Section IV-F), yet the physical experiments in Section IV-G do not compare against RRT, A*, or the unfine-tuned LLM, and no analysis is given for generalization to obstacle layouts outside the training distribution. In addition, the statement that 'RRT planner provides optimal shortest paths' is incorrect: RRT is not an optimal planner. The training-data claim should be corrected, and the physical validation needs a baseline comparison and an out-of-distribution test to support the conclusion that the fine-tuned model has robust spatial reasoning.
minor comments (7)
  1. [IV-D1] The phrase 'DeepSeek-R1 fails in all 10 trials, underscoring the limitations of purely reactive methods' mischaracterizes DeepSeek-R1, which is a large reasoning model, not a purely reactive method.
  2. [IV-B] The ESS definition ESS=SR/ACT with units %/s is unusual; please state explicitly how it normalizes for path length and why it is preferred over standard metrics.
  3. [III-A] The statement that a safety margin of 1.2 m is 'maintained around the obstacle along the edges' is ambiguous; specify whether the margin is measured from the obstacle surface or center.
  4. [IV-F, Table IV] Table IV lists Epochs=60 but the text says 'trained over 100 epochs' and Fig. 6 shows convergence after 60; please reconcile this inconsistency.
  5. [Abstract] The abstract says 'eight representative LLM and VLM approaches' but the paper actually evaluates eight LLMs and five VLMs; please rephrase to avoid undercounting the models.
  6. [IV-G] The statement that the model was 'requested from the Hugging Face server' suggests the LLM is not running fully onboard the Jetson; clarify the deployment split between onboard and remote inference and its effect on the real-time claim.
  7. [III-B, IV-A] The VLM simulation uses 1 m obstacles while the LLM simulation uses 2 m obstacles; please justify or align these settings so the two benchmarks are comparable.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the benchmark is external, the fine-tuning labels are transparently RRT-generated, and the physical flights are a genuine deployment test.

full rationale

We walked the paper's derivation chain. The central comparison (Table I) pits eight pretrained LLMs and three classic planners (A*, RRT, Q-learning) against identical coordinate-based prompts and the same Gazebo simulator; the LLM's waypoints are parsed and refined, but the success metric is determined by actual simulated flight, not by any quantity re-entering the model's input. No equation in Sections III-A, III-B, or IV-A reduces a prediction to a fitted parameter. The LoRA fine-tuning in Section IV-F uses 5,000 RRT-computed waypoint sequences as ground truth, and the physical experiments in Section IV-G test the resulting system on real flights with vision-derived obstacle positions. This is a standard supervised-learning pipeline rather than a circular one: RRT labels are external ground truth, the physical task adds perception noise not present in the labels, and success is judged by real collision-free flight. The paper's notable weaknesses—unvalidated YOLOv8+depth localization accuracy, single-shot physical trials, and the fact that any demonstrated 'spatial reasoning' is partly inherited from the RRT training generator—are correctness and generalization concerns, not definitional reductions. We found no load-bearing self-citation, no imported uniqueness theorem, and no ansatz smuggled in by citation. The derivation chain is therefore self-contained against external benchmarks, and the circularity score is 0.

Assumptions & free parameters 4 free parameters · 4 assumptions · 0 invented entities

The listed constants and assumptions are design choices that bound the validity of the safety and generalization claims. There are no hidden fitted constants beyond the fine-tuning hyperparameters and hand-set clearance/interpolation values.

free parameters (4)
  • Safety margin for obstacle waypoints = 1.2 m
    Hand-selected in Section III-A for the LLM path refinement module; collision-free success depends on it.
  • Interpolation step size = 0.5 m (LLM), 0.1 m (VLM)
    Hand-selected in Sections III-A and III-B; determines path smoothness and flight time.
  • Minimum clearance in synthetic training data = 1 m
    Used to generate RRT waypoints for the 5,000-entry fine-tuning dataset in Section IV-F; defines the target behavior of the model.
  • LoRA fine-tuning hyperparameters = lr 2e-4, batch size 4, 60-100 epochs, rank unspecified
    Chosen by empirical testing (Section IV-F, Table IV); no sensitivity analysis is reported.
assumptions (4)
  • domain assumption A prompt with start, goal, and obstacle coordinates is sufficient for an LLM to propose collision-free waypoints after interpolation.
    Assumed in Section III-A and validated only on the simple simulation scenarios; no formal or out-of-distribution guarantee.
  • domain assumption YOLOv8 and RGB-D depth provide obstacle locations accurate enough for the clearance margins.
    Invoked in Sections III-C and IV-G, but no localization error or calibration is reported.
  • domain assumption RRT-generated synthetic trajectories are a valid supervised target for real-world drone navigation.
    Training set in Section IV-F is produced by RRT; real-flight success is then used as evidence without quantifying the sim-to-real gap.
  • domain assumption The A* grid discretization is a fair baseline for the continuous coordinate task.
    Baseline comparison in Section IV-C discretizes the workspace at 0.12 m cells; this may handicap or favor baselines relative to the LLM's continuous waypoints.

how reviews work

0 comments
Cite this review

Pith. "Pith review of FM-Planner: Foundation Model Guided Path Planning for Autonomous Drone Navigation." pith.science (2026). https://pith.science/paper/R3OZIA66

@misc{pith2026250520783,
  author       = {Pith},
  title        = {Pith review of: FM-Planner: Foundation Model Guided Path Planning for Autonomous Drone Navigation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/R3OZIA66}},
  note         = {Machine review of arXiv:2505.20783}
}
read the original abstract

Path planning is a critical component in autonomous drone operations, enabling safe and efficient navigation through complex environments. Recent advances in foundation models, particularly large language models (LLMs) and vision-language models (VLMs), have opened new opportunities for enhanced perception and intelligent decision-making in robotics. However, their practical applicability and effectiveness in global path planning remain relatively unexplored. This paper proposes foundation model-guided path planners (FM-Planner) and presents a comprehensive benchmarking study and practical validation for drone path planning. Specifically, we first systematically evaluate eight representative LLM and VLM approaches using standardized simulation scenarios. To enable effective real-time navigation, we then design an integrated LLM-Vision planner that combines semantic reasoning with visual perception. Furthermore, we deploy and validate the proposed path planner through real-world experiments under multiple configurations. Our findings provide valuable insights into the strengths, limitations, and feasibility of deploying foundation models in real-world drone applications and providing practical implementations in autonomous flight. Project site: https://github.com/NTU-ICG/FM-Planner.

Figures

Figures reproduced from arXiv: 2505.20783 by the authors.

Figure 1
Figure 1. The real-world autonomous flight with a foundation model-guided [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The framework of the foundation model guided path planners with prompt and task descriptions. (a) The LLM-guided path planner with purely textual [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. System architecture of the LLM-Vision-guided path planner framework. A user-provided instruction is combined with real-time visual context from [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Input bird’s-eye view images for the VLM-guided path planner test. [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 6
Figure 6. Figure 6: The training curve of Llama-3.1-8B-Instruct during fine-tuning. [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 7
Figure 7. Figure 7: The configuration of physical experiments. (a) The configuration of QAV drone with the companion computer; (b) The autonomous flight area with [PITH_FULL_IMAGE:figures/full_fig_p008_7.png]
Figure 8
Figure 8. Figure 8: The real-world autonomous flight with the developed LLM-Vision planner under various tasks in Table [PITH_FULL_IMAGE:figures/full_fig_p008_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

36 extracted references · 24 canonical work pages

  1. [1]

    Collaborative target search with a visual drone swarm: An adaptive curriculum embedded multistage re- inforcement learning approach,

    J. Xiao, P. Pisutsin, and M. Feroskhan, “Collaborative target search with a visual drone swarm: An adaptive curriculum embedded multistage re- inforcement learning approach,”IEEE Transactions on Neural Networks and Learning Systems, 2023

    2023

  2. [2]

    Downside up: Rethinking parcel position for aerial delivery,

    P. M. Kornatowski, M. Feroskhan, W. J. Stewart, and D. Floreano, “Downside up: Rethinking parcel position for aerial delivery,”IEEE Robotics and Automation Letters, vol. 5, no. 3, pp. 4297–4304, 2020

    work page 2020

  3. [3]

    Industrial uav-based unsupervised domain adaptive crack recognitions: From database towards real-site infrastruc- tural inspections,

    K. Liu and B. M. Chen, “Industrial uav-based unsupervised domain adaptive crack recognitions: From database towards real-site infrastruc- tural inspections,”IEEE Transactions on Industrial Electronics, vol. 70, no. 9, pp. 9410–9420, 2023

    work page 2023

  4. [4]

    Multitarget assignment under uncer- tain information through decision support systems,

    Q. Jia, J. Xiao, and M. Feroskhan, “Multitarget assignment under uncer- tain information through decision support systems,”IEEE Transactions on Industrial Informatics, vol. 20, no. 8, pp. 10 636–10 646, 2024

    work page 2024

  5. [5]

    Uav path planning based on the combination of a-star algorithm and rrt-star algorithm,

    Q. Zhou and G. Liu, “Uav path planning based on the combination of a-star algorithm and rrt-star algorithm,” in2022 IEEE International Conference on Unmanned Systems (ICUS), 2022, pp. 146–151

    work page 2022

  6. [6]

    An efficient rrt-based framework for planning short and smooth wheeled robot motion under kinodynamic constraints,

    B. Hu, Z. Cao, and M. Zhou, “An efficient rrt-based framework for planning short and smooth wheeled robot motion under kinodynamic constraints,”IEEE Transactions on Industrial Electronics, vol. 68, no. 4, pp. 3292–3302, 2021

    work page 2021

  7. [7]

    Robust and effi- cient quadrotor trajectory generation for fast autonomous flight,

    B. Zhou, F. Gao, L. Wang, C. Liu, and S. Shen, “Robust and effi- cient quadrotor trajectory generation for fast autonomous flight,”IEEE Robotics and Automation Letters, vol. 4, no. 4, pp. 3529–3536, 2019

    2019

  8. [8]

    Ego-planner: An esdf- free gradient-based local planner for quadrotors,

    X. Zhou, Z. Wang, H. Ye, C. Xu, and F. Gao, “Ego-planner: An esdf- free gradient-based local planner for quadrotors,”IEEE Robotics and Automation Letters, vol. 6, no. 2, pp. 478–485, 2021

    2021

Show all 36 references
  1. [9]

    Vision-based learning for drones: A survey,

    J. Xiao, R. Zhang, Y . Zhang, and M. Feroskhan, “Vision-based learning for drones: A survey,”IEEE Transactions on Neural Networks and Learning Systems, pp. 1–21, 2025

    2025

  2. [10]

    Aggressive quadrotor flight using curiosity-driven reinforcement learning,

    Q. Sun, J. Fang, W. X. Zheng, and Y . Tang, “Aggressive quadrotor flight using curiosity-driven reinforcement learning,”IEEE Transactions on Industrial Electronics, vol. 69, no. 12, pp. 13 838–13 848, 2022

    2022

  3. [11]

    Learning multipursuit evasion for safe tar- geted navigation of drones,

    J. Xiao and M. Feroskhan, “Learning multipursuit evasion for safe tar- geted navigation of drones,”IEEE Transactions on Artificial Intelligence, vol. 5, no. 12, pp. 6210–6224, 2024

    2024

  4. [12]

    Learning cross-modal visuo- motor policies for autonomous drone navigation,

    Y . Zhang, J. Xiao, and M. Feroskhan, “Learning cross-modal visuo- motor policies for autonomous drone navigation,”IEEE Robotics and Automation Letters, vol. 10, no. 6, pp. 5425–5432, 2025

    2025

  5. [13]

    Champion-level drone racing using deep reinforcement learning,

    E. Kaufmann, L. Bauersfeld, A. Loquercio, M. M ¨uller, V . Koltun, and D. Scaramuzza, “Champion-level drone racing using deep reinforcement learning,”Nature, vol. 620, no. 7976, pp. 982–987, 2023

    2023

  6. [14]

    Multiagent path finding using deep reinforcement learning coupled with hot supervision contrastive loss,

    L. Chen, Y . Wang, Y . Mo, Z. Miao, H. Wang, M. Feng, and S. Wang, “Multiagent path finding using deep reinforcement learning coupled with hot supervision contrastive loss,”IEEE Transactions on Industrial Electronics, vol. 70, no. 7, pp. 7032–7040, 2023

    2023

  7. [15]

    An efficient reinforcement learning-based cooperative navigation algorithm for multiple uavs in complex environments,

    L. Zhang, W. Yi, H. Lin, J. Peng, and P. Gao, “An efficient reinforcement learning-based cooperative navigation algorithm for multiple uavs in complex environments,”IEEE Transactions on Industrial Informatics, vol. 20, no. 10, pp. 12 396–12 406, 2024

    2024

  8. [16]

    Toward collaborative multitar- get search and navigation with attention-enhanced local observation,

    J. Xiao, P. Pisutsin, and M. Feroskhan, “Toward collaborative multitar- get search and navigation with attention-enhanced local observation,” Advanced Intelligent Systems, p. 2300761, 2024

    2024

  9. [17]

    Deep reinforcement learning of collision-free flocking policies for multiple fixed-wing uavs using local situation maps,

    C. Yan, C. Wang, X. Xiang, Z. Lan, and Y . Jiang, “Deep reinforcement learning of collision-free flocking policies for multiple fixed-wing uavs using local situation maps,”IEEE Transactions on Industrial Informat- ics, vol. 18, no. 2, pp. 1260–1270, 2022

    2022

  10. [18]

    Learning resilient formation control of drones with graph attention network,

    J. Xiao, X. Fang, Q. Jia, and M. Feroskhan, “Learning resilient formation control of drones with graph attention network,”IEEE Internet of Things Journal, pp. 1–1, 2025

    2025

  11. [19]

    Gpt-4 technical report,

    J. Achiam, S. Adler, S. Agarwal, L. Ahmad, I. Akkaya, F. L. Aleman, D. Almeida, J. Altenschmidt, S. Altman, S. Anadkatet al., “Gpt-4 technical report,”arXiv preprint arXiv:2303.08774, 2023

    2023 arXiv

  12. [20]

    Flamingo: a visual language model for few-shot learning,

    J.-B. Alayrac, J. Donahue, P. Luc, A. Miech, I. Barr, Y . Hasson, K. Lenc, A. Mensch, K. Millican, M. Reynoldset al., “Flamingo: a visual language model for few-shot learning,”Advances in neural information processing systems, vol. 35, pp. 23 716–23 736, 2022

    2022

  13. [21]

    Pe-gpt: A new paradigm for power electronics design,

    F. Lin, X. Li, W. Lei, J. J. Rodriguez-Andina, J. M. Guerrero, C. Wen, X. Zhang, and H. Ma, “Pe-gpt: A new paradigm for power electronics design,”IEEE Transactions on Industrial Electronics, vol. 72, no. 4, pp. 3778–3791, 2025

    2025

  14. [22]

    Harnessing the power of large language model for effective web api recommendation,

    S. Qin, Y . Zhao, H. Wu, L. Zhang, and Q. He, “Harnessing the power of large language model for effective web api recommendation,”IEEE Transactions on Industrial Informatics, pp. 1–11, 2025

    2025

  15. [23]

    Large-scale visual language model boosted by contrast domain adaptation for intelligent industrial visual monitoring,

    H. Wang, C. Li, and Y .-F. Li, “Large-scale visual language model boosted by contrast domain adaptation for intelligent industrial visual monitoring,”IEEE Transactions on Industrial Informatics, vol. 20, no. 12, pp. 14 114–14 123, 2024

    2024

  16. [24]

    Leveraging large language models to empower bayesian networks for reliable human- robot collaborative disassembly sequence planning in remanufacturing,

    L. Xia, Y . Hu, J. Pang, X. Zhang, and C. Liu, “Leveraging large language models to empower bayesian networks for reliable human- robot collaborative disassembly sequence planning in remanufacturing,” IEEE Transactions on Industrial Informatics, vol. 21, no. 4, pp. 3117– 3126,...

    2025

  17. [25]

    Can foundation models perform zero-shot task specification for robot ma- nipulation?

    Y . Cui, S. Niekum, A. Gupta, V . Kumar, and A. Rajeswaran, “Can foundation models perform zero-shot task specification for robot ma- nipulation?” inLearning for dynamics and control conference. PMLR, 2022, pp. 893–905

    2022

  18. [26]

    Towards realistic uav vision-language navigation: Platform, benchmark, and methodology,

    X. Wang, D. Yang, Z. Wang, H. Kwan, J. Chen, W. Wu, H. Li, Y . Liao, and S. Liu, “Towards realistic uav vision-language navigation: Platform, benchmark, and methodology,”arXiv preprint arXiv:2410.07087, 2024

    2024 arXiv

  19. [27]

    Foundation models in robotics: Applications, challenges, and the future,

    R. Firoozi, J. Tucker, S. Tian, A. Majumdar, J. Sun, W. Liu, Y . Zhu, S. Song, A. Kapoor, K. Hausmanet al., “Foundation models in robotics: Applications, challenges, and the future,”The International Journal of Robotics Research, p. 02783649241281508, 2023

    2023

  20. [28]

    Leviosa: Natural language-based uncrewed aerial vehicle trajectory generation,

    G. Aikins, M. P. Dao, K. J. Moukpe, T. C. Eskridge, and K.-D. Nguyen, “Leviosa: Natural language-based uncrewed aerial vehicle trajectory generation,”Electronics, vol. 13, no. 22, p. 4508, 2024

    2024

  21. [29]

    A formal basis for the heuristic determination of minimum cost paths,

    P. E. Hart, N. J. Nilsson, and B. Raphael, “A formal basis for the heuristic determination of minimum cost paths,”IEEE transactions on Systems Science and Cybernetics, vol. 4, no. 2, pp. 100–107, 1968

    1968

  22. [30]

    Sampling-based algorithms for optimal motion planning,

    S. Karaman and E. Frazzoli, “Sampling-based algorithms for optimal motion planning,”The international journal of robotics research, vol. 30, no. 7, pp. 846–894, 2011

    2011

  23. [31]

    Resilient real-time decision-making for autonomous mobile robot path planning in complex dynamic environments,

    X. Hai, Z. Zhu, Y . Liu, A. W. H. Khong, and C. Wen, “Resilient real-time decision-making for autonomous mobile robot path planning in complex dynamic environments,”IEEE Transactions on Industrial Electronics, pp. 1–12, 2025

    2025

  24. [32]

    Deep drone acrobatics,

    E. Kaufmann, A. Loquercio, R. Ranftl, M. M ¨uller, V . Koltun, and D. Scaramuzza, “Deep drone acrobatics,”Proceedings of Robotics: Science and Systems XVI, 2020

    2020

  25. [33]

    An adaptive conversion speed q-learning algorithm for search and rescue uav path planning in unknown environments,

    J. Wu, Y . Sun, D. Li, J. Shi, X. Li, L. Gao, L. Yu, G. Han, and J. Wu, “An adaptive conversion speed q-learning algorithm for search and rescue uav path planning in unknown environments,”IEEE Transactions on Vehicular Technology, vol. 72, no. 12, pp. 15 391–15 404, 2023

    2023

  26. [34]

    Federated imitation learning for uav swarm coordination in urban traffic monitoring,

    B. Yang, H. Shi, and X. Xia, “Federated imitation learning for uav swarm coordination in urban traffic monitoring,”IEEE Transactions on Industrial Informatics, vol. 19, no. 4, pp. 6037–6046, 2023

    2023

  27. [35]

    Npe-drl: Enhancing perception constrained obstacle avoidance with non-expert policy guided reinforcement learning,

    Y . Zhang, C. Yan, J. Xiao, and M. Feroskhan, “Npe-drl: Enhancing perception constrained obstacle avoidance with non-expert policy guided reinforcement learning,”IEEE Transactions on Artificial Intelligence, 2024

    2024

  28. [36]

    On the opportunities and risks of foundation models,

    R. Bommasani, D. A. Hudson, E. Adeli, R. Altman, and et al., “On the opportunities and risks of foundation models,”arXiv preprint arXiv:2108.07258, 2021

    2021 arXiv

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.