Enhancing Robot Navigation Policies with Task-Specific Uncertainty Managements

Gokul Puthumanaillam; Jose Fuentes; Leonardo Bobadilla; Melkior Ornik; Paulo Padrao

REVIEW 3 major objections 4 minor 45 references

Enhancing Robot Navigation Policies with Task-Specific Uncertainty Managements

T0 review · 3 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read Navigation policies conditioned on task-specific uncertainty maps outperform six baselines in real-world lake navigation.

desk verdict The TSUM idea is genuinely new and coherent, but the paper's headline performance claim is undercut by a missing experimental protocol and a likely feature-availability confound. read the letter →

arxiv 2505.13837 v1 pith:YLL56XEU submitted 2025-05-20 cs.RO cs.AIcs.LG

Gokul Puthumanaillam , Paulo Padrao , Jose Fuentes , Leonardo Bobadilla , Melkior Ornik This is my paper · ORCID

classification cs.ROcs.AIcs.LG

keywords task-specificuncertaintymapsuncertainty-guidednavigationsoftactor-criticreinforcementlearningforautonomoussurfacevehicleactivelocalizationnaturallanguagetaskspecification

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

This paper argues that robot navigation policies should manage uncertainty according to what each task actually requires at each location, rather than minimizing uncertainty everywhere or applying one global threshold. To that end it introduces Task-Specific Uncertainty Maps (TSUMs), scalar fields that assign an acceptable uncertainty level to every location from a natural-language task description plus environmental constraints. The paper claims that conditioning a Soft Actor-Critic policy on the TSUM and the robot's current uncertainty, in a variant it calls G-SAC, outperforms six baselines on all four tested autonomous-surface-vehicle tasks in both task completion rate and reward. A sympathetic reader would care because the method offers a way to encode task constraints without hand-designed reward terms.

What carries the argument

The load-bearing object is the Task-Specific Uncertainty Map, $U_\tau(l) = w_\Phi \Phi_\tau(l) + w_C C_\tau(l) + w_E E(l)$, a scalar per location combining task relevance, constraints such as safety restrictions, and environmental factors. Task semantics are extracted from the natural-language specification with a RoBERTa-based parser; spatial embeddings are aligned to these semantics with a triplet loss and an attention mechanism. The TSUM enters the policy through an augmented state $\tilde{s} = [s, U_\tau(s), u(s)]$, and a Soft Actor-Critic variant, G-SAC, learns to act on it. The mechanism carries the argument by replacing hand-tuned uncertainty penalties with a spatially varying, task-derived tolerance signal.

What would settle it

Train a baseline that receives the same augmented state as G-SAC (location, acceptable uncertainty, current uncertainty) but whose policy is prevented from using the acceptable-uncertainty component; if it matches G-SAC's task completion rate, the benefit is not from task-specific mapping. Alternatively, run G-SAC with a spatially uniform TSUM; high performance under that condition would show the spatial structure is not the cause.

Watch

Extended reading notes

Core claim

The central discovery is that a navigation policy can profit from being explicitly told how much uncertainty is acceptable at its current location. Formally, the state is augmented to $\tilde{s} = [s, U_\tau(s), u(s)]$, where $U_\tau(s)$ is the TSUM value at the robot's location and $u(s)$ is the current state-estimation uncertainty. With this augmentation, G-SAC learns to request precise GPS localization only where the TSUM demands tight tolerances, and tolerates noisy estimation in open areas. In the reported experiments, G-SAC achieves the highest Task Completion Rate and average reward in all four tasks, with the largest gains over the handcrafted policy (HEU) that switches to GPS near obstacles.

Load-bearing premise

The load-bearing premise is that the comparison baselines receive the same environment observations and cost structure as G-SAC; if those baselines are not also given the robot's current uncertainty value, the reported advantage could come from that extra input rather than from task-specific uncertainty management.

Editorial extensions

If this is right

If G-SAC's advantage holds, navigation systems can trade expensive precise localization for cheap noisy estimation in regions where the task tolerates it, reducing operational cost without sacrificing task success.
A single trained policy could be redirected to a new task by recomputing the TSUM, without re-engineering the reward or retraining the uncertainty management logic.
Natural-language task descriptions become a usable input to low-level navigation policies, since TSUMs translate those descriptions into location-wise constraints.
The approach transfers to any robot with adjustable sensing or state-estimation quality, not just surface vehicles, whenever a task can be described as spatially varying precision needs.

Reading between the lines

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

A stricter test than the one reported would hold the augmented state fixed and vary only the TSUM: two tasks sharing a map should yield identical behavior, and two maps for the same task should yield different behavior; the current experiments do not include this control, so the causal role of the map is inferred rather than isolated.
The method inherits the reliability of its language and embedding components; an ambiguous or misparsed task description would produce a map that may not reflect the operator's intent, making end-to-end testing on natural-language input a natural next experiment.
The paper's framing suggests a spatial generalization: TSUMs could be extended from a scalar per location to vector-valued tolerances encoding multiple uncertainty sources, such as pose, velocity, and heading, with coupled spatial constraints.

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.

Desk Editor's Note

read the letter

The new thing here is the TSUM concept: a per-location acceptable-uncertainty map derived from a natural-language task specification, fed into a SAC policy as part of the augmented state. That combination—language-conditioned spatial requirements plus policy conditioning—is not in the cited literature, and the idea is sensible. If a robot knows where precision matters and how uncertain it currently is, it can learn to request costly GPS only where the task demands it. The method description is clear, and the framing as an alternative to uniform uncertainty penalties is fair.

Where the paper falls short is the evidence. Table 1 reports that G-SAC beats six baselines by large margins, but there are no error bars, no number of seeds, no significance tests, and no protocol description. That alone would make me treat the numbers as preliminary. The bigger problem is the stress-test confound: G-SAC's state is [s, U_tau(s), u(s)], and the text only says baselines share 'environment observations and cost structure,' not that they receive u(s). If the baselines do not observe the robot's current state-estimation uncertainty, then G-SAC has an extra input that directly indicates when to request GPS, and the reported gains could come from that alone rather than from the task-specific map. The paper also leaves the TSUM weights (w_Phi, w_C, w_E) undisclosed, and the abstract's 'significant performance gains' from real-world tests are unsupported by any trial counts or statistical procedures.

None of this kills the idea. The concept is independent of the experimental weaknesses, and the 3-page format may simply not allow for full details. But as submitted, the central empirical claim is not yet separable from the confound. A full version with information-preserving baselines, variance reports, and a proper ablation (G-SAC with and without u(s)) would be a solid contribution to active localization and uncertainty-aware navigation.

I would bring this to a reading group as a 'here's a promising idea with a cautionary tale about evaluation' example, but I would not cite the quantitative results. I would still send it to peer review—a serious referee could help the authors see exactly what is missing, and the idea deserves engagement. Just be clear that the experiments need major revision before the performance claims can be trusted.

Referee Report

3 major / 4 minor

Summary. The paper introduces GUIDE (Generalized Uncertainty Integration for Decision-Making and Execution), a framework that encodes task-specific acceptable uncertainty at each location through Task-Specific Uncertainty Maps (TSUMs). The TSUM is computed as a weighted combination of task-relevance, constraint, and environment factors, with semantics extracted using a RoBERTa-based parser and aligned to spatial embeddings via triplet loss and attention. The framework is instantiated as G-SAC, a Soft Actor-Critic variant whose augmented state includes both the TSUM value and the current state-estimation uncertainty. The manuscript reports that G-SAC achieves the highest Task Completion Rate (TCR) and reward over six baselines across four navigation tasks in Table 1, and the abstract claims significant performance gains in real-world tests.

Significance. If the central claim were established, the paper would make a useful contribution: it addresses a real problem in active localization and uncertainty-aware navigation, where the cost of precision should depend on spatial context. The TSUM idea is conceptually coherent and the policy-conditioning mechanism is plausible, with the potential to reduce manual reward shaping. The empirical setup, using an autonomous surface vehicle with optional precise GPS at a penalty, is well suited to demonstrating the value of task-specific uncertainty management. The paper also compares against a reasonable range of baselines, including penalized, bootstrapped, risk-aware, and handcrafted alternatives. However, the current evidence is not sufficient to support the central claim, because the comparison may be confounded by an information asymmetry and because the reported table lacks statistical support. The framework is potentially significant, but the manuscript needs a fairer and more rigorous empirical evaluation before the claims can be accepted.

major comments (3)

[Section 2, 'Policy Conditioning'; Section 3, 'Baselines and Ablations'] The central empirical claim is confounded by a potential feature-availability asymmetry. Section 2 defines the augmented state as ~s = [s, U_τ(s), u(s)] and states that the G-SAC policy and Q-function receive this augmented state. Section 3 says only that all baselines 'share the same environment observations and cost structure,' and it does not state whether SAC, SAC-P, B-SAC, CVaR, RAA, or HEU also observe u(s), the robot's current state-estimation uncertainty. If the baselines do not receive u(s), then the reported 10-40 point TCR advantage of G-SAC could come from direct access to this extra state feature, independent of the TSUM contribution. The authors should specify exactly which information each baseline receives, and ideally add ablations such as G-SAC without u(s) and SAC with u(s), so that the TSUM's contribution is isolated.
[Section 3, Table 1 and the abstract] Table 1 reports TCR and reward values with no number of episodes, number of seeds, variance, error bars, or significance tests. The abstract's phrase 'significant performance gains' is therefore unsupported in the statistical sense. For a comparison over four tasks and seven methods, the absence of any measure of variability makes it impossible to assess whether the reported margins are meaningful, and the abstract should not claim significance without such evidence. The authors should report mean and standard deviation across multiple independent runs and use an appropriate significance test or overlap analysis, and they should clarify whether Table 1 reports simulated or real-world results.
[Section 2, 'Task-Specific Uncertainty Maps (TSUM)'] The TSUM formula U_τ(l) = w_Φ Φ_τ(l) + w_C C_τ(l) + w_E E(l) introduces three free weights, and the experiments also include a GPS request penalty, yet the manuscript does not report the values of these weights, how they were chosen, or any sensitivity analysis. This matters for two reasons. First, without the weight values the experiments cannot be reproduced. Second, the paper claims the framework avoids 'extensive reward engineering' and 'ad hoc penalty terms,' but the TSUM weights themselves appear to require manual tuning; the authors should either provide the values and tuning procedure or temper the claim about eliminating reward engineering.

minor comments (4)

[Section 3, Table 1] The task heading 'A void' appears to be a typo for 'Avoid.'
[References, reference [20]] The name 'Thoams J Fuchs' appears to be a typo for 'Thomas J Fuchs.'
[Throughout] The capitalization of 'GUIDEd SAC' is inconsistent; the text uses 'GUIDEd' in Section 2 and 'G-SAC' in Section 3, and one consistent form should be used.
[Section 2, 'Task-Specific Uncertainty Maps (TSUM)'] The notation switches between location l in the definition of U_τ(l) and state s in the augmented state ~s = [s, U_τ(s), u(s)]; the authors should clarify the relationship between l and s, especially whether U_τ is evaluated at the estimated robot position.

Circularity Check

0 steps flagged · score 2.0 of 10

No circular derivation: G-SAC's TSUM conditioning is an input representation tested empirically; the reported advantage is an experimental comparability question, not a reduction to the paper's own inputs.

full rationale

The paper's core claim is empirical: conditioning SAC on a Task-Specific Uncertainty Map (TSUM) and current uncertainty improves task completion and reward in four navigation tasks. The TSUM is defined as U_tau(l) = w_Phi Phi_tau(l) + w_C C_tau(l) + w_E E(l), assembled from a RoBERTa task parser, spatial embeddings, triplet loss, and attention, and is then fed as an extra input to G-SAC. No equation constructs G-SAC's performance from the TSUM by definition; the policy is trained with SAC and evaluated in simulation/real-world trials against six baselines. The same-author citations ([23], [24], [25], [27]) support general RL effectiveness, a categorization of uniform trade-off methods, and the GPS-request experimental mechanism, but none is invoked as a uniqueness theorem or as the source of the reported superiority, so they are not load-bearing. The most serious weakness is a possible feature-availability confound: Section 2 defines the augmented state as [s, U_tau(s), u(s)], while Section 3 only states that baselines 'share the same environment observations and cost structure'; if baselines do not observe u(s), the comparison may reflect an extra state feature rather than task-specific uncertainty management. That is an experimental fairness issue, not circularity, because no fitted parameter is renamed as a prediction and no result reduces by construction to its input. The abstract's 'significant performance gains' also lack reported variance or significance tests, but that is a reporting limitation, not a circular step.

Assumptions & free parameters 4 free parameters · 3 assumptions · 1 invented entities

The central method relies on a postulated scalar map formula with undisclosed weights, on the assumption that language can be mapped to spatial uncertainty needs, and on access to the robot's current uncertainty. No external validation of the map is provided; the map is only used in the same pipeline it is claimed to improve.

free parameters (4)

w_Phi (task relevance weight)
Weight on semantic relevance in U_tau; value not specified, likely hand-chosen; no sensitivity study.
w_C (constraint weight)
Weight on safety and legal constraints in U_tau; value not reported; no sensitivity study.
w_E (environment factor weight)
Weight on environmental factors in U_tau; value not reported; no sensitivity study.
GPS request penalty
Cost penalty for requesting precise GPS; not reported. It affects the reward trade-off in all methods, so its value influences the comparison.

assumptions (3)

ad hoc to paper The acceptable uncertainty at a location is accurately represented by the scalar U_tau formula given in Section 2.
The whole method rests on the postulated weighted sum of relevance, constraints, and environmental factors, with no derivation or comparison to alternative representations.
domain assumption The RoBERTa parser and triplet-loss alignment map natural-language tasks to meaningful spatial uncertainty requirements.
Section 2 states that task semantics are extracted and aligned, but no validation is provided that the learned embedding similarity reflects true uncertainty needs.
domain assumption The robot has access to a reliable estimate of its own localization uncertainty u(s) at each state.
The augmented state includes u(s), but the paper does not say how this estimate is obtained or how accurate it is; without it G-SAC cannot condition as claimed.

invented entities (1)

Task-Specific Uncertainty Map (TSUM) U_tau(l)
purpose: Encodes acceptable uncertainty at each location for a language-specified task and is used as an extra conditioning input to the RL policy.
The central new representation; it is only validated through the same GUIDE policy, so its usefulness has no independent handle in this manuscript.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Enhancing Robot Navigation Policies with Task-Specific Uncertainty Managements." pith.science (2026). https://pith.science/paper/YLL56XEU

@misc{pith2026250513837,
  author       = {Pith},
  title        = {Pith review of: Enhancing Robot Navigation Policies with Task-Specific Uncertainty Managements},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/YLL56XEU}},
  note         = {Machine review of arXiv:2505.13837}
}

read the original abstract

Robots navigating complex environments must manage uncertainty from sensor noise, environmental changes, and incomplete information, with different tasks requiring varying levels of precision in different areas. For example, precise localization may be crucial near obstacles but less critical in open spaces. We present GUIDE (Generalized Uncertainty Integration for Decision-Making and Execution), a framework that integrates these task-specific requirements into navigation policies via Task-Specific Uncertainty Maps (TSUMs). By assigning acceptable uncertainty levels to different locations, TSUMs enable robots to adapt uncertainty management based on context. When combined with reinforcement learning, GUIDE learns policies that balance task completion and uncertainty management without extensive reward engineering. Real-world tests show significant performance gains over methods lacking task-specific uncertainty awareness.

Figures

Figures reproduced from arXiv: 2505.13837 by the authors.

**Figure 1.** During pretraining, semantic and spatial embeddings are aligned via triplet loss and attention. At deployment, TSUMs [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

45 extracted references · 37 canonical work pages

[1]

Mohammad Alali and Mahdi Imani. 2024. Bayesian reinforcement learning for navigation planning in unknown environments. Frontiers in Artificial Intelligence (2024)

work page 2024
[2]

Matthew Budd, Paul Duckworth, Nick Hawes, and Bruno Lacerda. 2023. Bayesian reinforcement learning for single-episode missions in partially unknown envi- ronments. In Conference on Robot Learning

work page 2023
[3]

Matthew Chang, Theophile Gervet, Mukul Khanna, Sriram Yenamandra, Dhruv Shah, So Yeon Min, Kavit Shah, Chris Paxton, Saurabh Gupta, Dhruv Batra, et al

work page
[4]

Tao Chen, Saurabh Gupta, and Abhinav Gupta. 2019. Learning exploration policies for navigation. arXiv preprint arXiv:1903.01959 (2019)

arXiv 2019
[5]

Gerald Cook and Feitian Zhang. 2020. Mobile robots: Navigation, control and sensing, surface robots and AUVs . John Wiley & Sons

work page 2020
[6]

Aidan Curtis, George Matheos, Nishad Gothoskar, Vikash Mansinghka, Joshua Tenenbaum, Tomás Lozano-Pérez, and Leslie Pack Kaelbling. 2024. Partially Observable Task and Motion Planning with Uncertainty and Risk Awareness. arXiv preprint arXiv:2403.10454 (2024)

arXiv 2024
[7]

Dieter Fox, Sebastian Thrun, Wolfram Burgard, and Frank Dellaert. 2001. Particle filters for mobile robot localization. InSequential Monte Carlo methods in practice

work page 2001
[8]

Alejandro Gonzalez-Garcia and Herman Castañeda. 2021. Guidance and control based on adaptive sliding mode strategy for a USV subject to uncertainties. IEEE Journal of Oceanic Engineering (2021)

work page 2021

Show all 45 references

[9]

Faiza Gul, Wan Rahiman, and Syed Sahal Nazli Alhady. 2019. A comprehensive study for robot navigation techniques. Cogent Engineering (2019)

2019
[10]

Siyu Guo, Xiuguo Zhang, Yiquan Du, Yisong Zheng, and Zhiying Cao. 2021. Path planning of coastal ships based on optimized DQN reward function. Journal of Marine Science and Engineering (2021)

2021
[11]

Fredrik Gustafsson, Fredrik Gunnarsson, Niclas Bergman, Urban Forssell, Jonas Jansson, Rickard Karlsson, and P-J Nordlund. 2002. Particle filters for positioning, navigation, and tracking. IEEE Transactions on signal processing (2002)

2002
[12]

Tuomas Haarnoja, Aurick Zhou, Pieter Abbeel, and Sergey Levine. 2018. Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor. In International Conference on Machine Learning

2018
[13]

Astghik Hakobyan, Gyeong Chan Kim, and Insoon Yang. 2019. Risk-aware motion planning and control using CVaR-constrained optimization. IEEE Robotics and Automation letters (2019)

2019
[14]

Jonathan Ko and Dieter Fox. 2009. GP-BayesFilters: Bayesian filtering using Gaussian process prediction and observation models. Autonomous Robots (2009)

2009
[15]

Aviral Kumar, Justin Fu, Matthew Soh, George Tucker, and Sergey Levine. 2019. Stabilizing off-policy Q-Learning via bootstrapping error reduction. Advances in Neural Information Processing Systems (2019)

2019
[16]

Guofa Li, Yifan Yang, Shen Li, Xingda Qu, Nengchao Lyu, and Shengbo Eben Li
[17]

Yueming Li, Mingquan Ma, Jian Cao, Guobin Luo, Depeng Wang, and Weiqiang Chen. 2024. A Method for Multi-AUV Cooperative Area Search in Unknown Environment Based on Reinforcement Learning. Journal of Marine Science and Engineering (2024)

2024
[18]

Yinhan Liu. 2019. Roberta: A robustly optimized BERT pretraining approach. arXiv preprint arXiv:1907.11692 (2019)

2019 arXiv
[19]

Steven Morad, Ryan Kortvelesy, Matteo Bettini, Stephan Liwicki, and Amanda Prorok. 2023. POPGym: Benchmarking partially observable reinforcement learn- ing. arXiv preprint arXiv:2303.01859 (2023)

2023 arXiv
[20]

Masahiro Ono, Thoams J Fuchs, Amanda Steffy, Mark Maimone, and Jeng Yen
[21]

Abhishek Padalkar, Acorn Pooley, Ajinkya Jain, Alex Bewley, Alex Herzog, Alex Irpan, Alexander Khazatsky, Anant Rai, Anikait Singh, Anthony Brohan, et al
[22]

BK Patle, Anish Pandey, DRK Parhi, AJDT Jagadeesh, et al. 2019. A review: On path planning strategies for navigation of mobile robot. Defence Technology (2019)

2019
[23]

Gokul Puthumanaillam, Xiangyu Liu, Negar Mehr, and Melkior Ornik. 2024. Weathering ongoing uncertainty: Learning and planning in a time-varying par- tially observable environment. In 2024 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 4612–4618

2024
[24]

Gokul Puthumanaillam, Jae Hyuk Song, Nurzhan Yesmagambet, Shinkyu Park, and Melkior Ornik. 2024. TAB-Fields: A Maximum Entropy Framework for Mission-Aware Adversarial Planning. arXiv preprint arXiv:2412.02570 (2024)

2024 arXiv
[25]

arXiv preprint arXiv:2310.08864 (2023)

Open X-Embodiment: Robotic learning datasets and RT-X models. arXiv preprint arXiv:2310.08864 (2023)

2023 arXiv
[26]

Krishan Rana, Ben Talbot, Vibhavari Dasagi, Michael Milford, and Niko Sün- derhauf. 2020. Residual reactive navigation: Combining classical and learned navigation strategies for deployment in unknown environments. In 2020 IEEE International Conference on Robotics and Automation

2020
[27]

Cesar A Rojas, Paulo Padrão, Jose Fuentes, Gregory M Reis, Arif R Albayrak, Batuhan Osmanoglu, and Leonardo Bobadilla. 2024. Combining multi-satellite remote and in-situ sensing for unmanned underwater vehicle state estimation. Ocean Engineering (2024)

2024
[28]

Dhruv Shah, Błażej Osiński, Sergey Levine, et al. 2023. LM-Nav: Robotic naviga- tion with large pre-trained models of language, vision, and action. In Conference on Robot Learning

2023
[29]

Gokul Puthumanaillam, Manav Vora, and Melkior Ornik. 2024. ComTraQ-MPC: Meta-Trained DQN-MPC Integration for Trajectory Tracking with Limited Active Localization Updates. arXiv preprint arXiv:2403.01564 (2024)

2024 arXiv
[30]

Matthijs TJ Spaan, Tiago S Veiga, and Pedro U Lima. 2015. Decision-theoretic planning under uncertainty with information rewards for active cooperative perception. Autonomous Agents and Multi-Agent Systems (2015)

2015
[31]

Krishnan Srinivasan, Benjamin Eysenbach, Sehoon Ha, Jie Tan, and Chelsea Finn. 2020. Learning to be safe: Deep RL with a safety critic. arXiv preprint arXiv:2010.14603 (2020)

2020 arXiv
[32]

Annalisa T Taylor, Thomas A Berrueta, and Todd D Murphey. 2021. Active learning in robotics: A review of control principles. Mechatronics (2021)

2021
[33]

Siddharth Singi, Zhanpeng He, Alvin Pan, Sandip Patel, Gunnar A Sigurdsson, Robinson Piramuthu, Shuran Song, and Matei Ciocarlie. 2024. Decision mak- ing for human-in-the-loop robotic agents via uncertainty-aware reinforcement learning. In 2024 IEEE International Conference on...

2024
[34]

Ning Wang, Yabiao Wang, Yuming Zhao, Yong Wang, and Zhigang Li. 2022. Sim-to-real: Mapless navigation for USVs using deep reinforcement learning. Journal of Marine Science and Engineering (2022)

2022
[35]

Zixiang Wang, Hao Yan, Zhuoyue Wang, Zhengjia Xu, Zhizhong Wu, and Yining Wang. 2024. Research on autonomous robots navigation based on reinforce- ment learning. In International Conference on Robotics, Artificial Intelligence and Intelligent Control. IEEE

2024
[36]

Zichao Yang, Diyi Yang, Chris Dyer, Xiaodong He, Alex Smola, and Eduard Hovy
[37]

Sebastian Thrun. 2002. Particle Filters in Robotics. In UAI

2002
[38]

Yu Zhang, Sarath Sreedharan, Anagha Kulkarni, Tathagata Chakraborti, Hankz Hankui Zhuo, and Subbarao Kambhampati. 2016. Plan explicability for robot task planning. In RSS Workshop on Planning for Human-Robot Interaction: Shared Autonomy and Collaborative Robotics

2016
[39]

Kai Zhu and Tao Zhang. 2021. Deep reinforcement learning based mobile robot navigation: A review. Tsinghua Science and Technology (2021)

2021
[40]

Yuanyang Zhu, Zhi Wang, Chunlin Chen, and Daoyi Dong. 2021. Rule-based reinforcement learning for efficient robot navigation with space reduction. IEEE/ASME Transactions on Mechatronics (2021)

2021
[42]

Denis Yarats, Amy Zhang, Ilya Kostrikov, Brandon Amos, Joelle Pineau, and Rob Fergus. 2021. Improving sample efficiency in model-free reinforcement learning from images. In AAAI Conference on Artificial Intelligence

2021
[2015]

In IEEE Aerospace Conference

Risk-aware planetary rover operation: Autonomous terrain classification and path planning. In IEEE Aerospace Conference
[2016]

In Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

Hierarchical attention networks for document classification. In Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
[2022]

Transportation research part C: emerging technologies (2022)

Decision making of autonomous vehicles in lane change scenarios: Deep reinforcement learning approaches with risk awareness. Transportation research part C: emerging technologies (2022)

2022
[2023]

arXiv preprint arXiv:2311.06430 (2023)

GOAT: Go to any thing. arXiv preprint arXiv:2311.06430 (2023)

2023 arXiv

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.

[1] [1]

Mohammad Alali and Mahdi Imani. 2024. Bayesian reinforcement learning for navigation planning in unknown environments. Frontiers in Artificial Intelligence (2024)

work page 2024

[2] [2]

Matthew Budd, Paul Duckworth, Nick Hawes, and Bruno Lacerda. 2023. Bayesian reinforcement learning for single-episode missions in partially unknown envi- ronments. In Conference on Robot Learning

work page 2023

[3] [3]

Matthew Chang, Theophile Gervet, Mukul Khanna, Sriram Yenamandra, Dhruv Shah, So Yeon Min, Kavit Shah, Chris Paxton, Saurabh Gupta, Dhruv Batra, et al

work page

[4] [4]

Tao Chen, Saurabh Gupta, and Abhinav Gupta. 2019. Learning exploration policies for navigation. arXiv preprint arXiv:1903.01959 (2019)

arXiv 2019

[5] [5]

Gerald Cook and Feitian Zhang. 2020. Mobile robots: Navigation, control and sensing, surface robots and AUVs . John Wiley & Sons

work page 2020

[6] [6]

Aidan Curtis, George Matheos, Nishad Gothoskar, Vikash Mansinghka, Joshua Tenenbaum, Tomás Lozano-Pérez, and Leslie Pack Kaelbling. 2024. Partially Observable Task and Motion Planning with Uncertainty and Risk Awareness. arXiv preprint arXiv:2403.10454 (2024)

arXiv 2024

[7] [7]

Dieter Fox, Sebastian Thrun, Wolfram Burgard, and Frank Dellaert. 2001. Particle filters for mobile robot localization. InSequential Monte Carlo methods in practice

work page 2001

[8] [8]

Alejandro Gonzalez-Garcia and Herman Castañeda. 2021. Guidance and control based on adaptive sliding mode strategy for a USV subject to uncertainties. IEEE Journal of Oceanic Engineering (2021)

work page 2021