pith. sign in

arxiv: 2604.24907 · v1 · submitted 2026-04-27 · 💻 cs.LO · cs.RO

Logic of Fuzzy Paths

Kush Grover , Pratham Gupta , Jan K\v{r}et\'insk\'y This is my paper

Pith reviewed 2026-05-07 17:41 UTC · model grok-4.3

classification 💻 cs.LO cs.RO
keywords temporal logicmotion planningsignal temporal logicfuzzy constraintsspecification learningroboticspath planningverification
0
0 comments X

The pith

A new temporal logic treats paths as first-class objects to separate geometry from logic in motion planning specifications.

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

The paper introduces a family of temporal logics for robot motion planning that builds on signal temporal logic but elevates entire paths to first-class status within the logic. This design choice decouples the geometric description of the path from the logical constraints applied to it. The resulting formulas are simpler and more readable for human users, while the satisfaction relation becomes more refined and can encode preferences among different valid behaviors. The logic achieves this through fuzzy, time-varying signal constraints. The authors demonstrate the approach on examples, provide a learning algorithm for deriving specifications from demonstrations, and discuss its use in model checking and monitoring.

Core claim

The logic of fuzzy paths operates on paths as primary objects rather than signals, using fuzzy time-varying constraints to define quantitative satisfaction. This separation yields simpler, more understandable formulas than those in signal temporal logic and supports a refined satisfaction notion that reflects preferences over behaviors. The framework improves usability for manual specification in traditional verification and enables more effective learning of tasks from demonstration data for controller synthesis.

What carries the argument

Paths treated as first-class citizens combined with fuzzy, time-varying signal constraints.

If this is right

  • Human engineers gain simpler and more intuitive ways to write motion planning specifications for robots.
  • Specifications become easier to learn automatically from sets of demonstration trajectories.
  • Satisfaction values can distinguish among multiple behaviors that all satisfy the specification according to preference.
  • The same framework supports both traditional model checking for verification and runtime monitoring.
  • The approach applies across multiple motion planning scenarios with flexibility in how constraints are defined.

Where Pith is reading between the lines

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

  • The path-centric view could allow geometric planners to generate candidate paths independently before logical constraints are applied.
  • Learned preferences might support controller synthesis that adapts to individual human operators in shared workspaces.
  • The fuzzy constraints may naturally extend to handle uncertainty in sensor data during planning.
  • Similar separation of concerns could be tested in related domains such as autonomous vehicle trajectory planning.

Load-bearing premise

That elevating paths to first-class objects and applying fuzzy constraints will produce simpler formulas and better learnability without adding new complexity to the semantics or monitoring algorithms.

What would settle it

A side-by-side comparison in which human-written specifications or learned formulas in the new logic are measured for length, readability, and accuracy against equivalent signal temporal logic formulas on the same motion planning tasks.

Figures

Figures reproduced from arXiv: 2604.24907 by Jan K\v{r}et\'insk\'y, Kush Grover, Pratham Gupta.

Figure 1
Figure 1. Figure 1: The running example of avoiding an obstacle in a view at source ↗
Figure 2
Figure 2. Figure 2: The “boxing” approach of specifying continuous view at source ↗
Figure 3
Figure 3. Figure 3: More (darker blue) and less (light blue) preferred view at source ↗
Figure 4
Figure 4. Figure 4: Real, observed paths are drawn in blue, the learned view at source ↗
Figure 5
Figure 5. Figure 5: The atom from Example 4.2 Example 4.2. Again, for our running example, we use two signals 𝑥 and 𝑦, denoting the robot’s position. Consider the mean 𝝁(𝑡) =  𝑡 𝑡/3  , the covariance matrix 𝝈 (𝑡) =  1/16 0 0 (1/2 − 𝑡/12) 2  , and the atom 𝝅 [0,3] 1 (𝑡) = ⟨𝝁, 𝝈⟩ [0,3] . This atom is shown in view at source ↗
Figure 6
Figure 6. Figure 6: Distance of the points (0, −1), (1.5, −1), and (3, −1) from the point-atoms 𝝅 [0,3] 1 (0), 𝝅 [0,3] 1 (1.5), and 𝝅 [0,3] 1 (3) respec￾tively where 𝝅 [0,3] 1 is the atom from Example 4.2. Example 4.5. We look at how semantics behave for the trajectory 𝜁 [0,3] (𝑡) =  𝑡 −1  w.r.t. the atom from Example 4.2 at 𝑡 = 0, 𝑡 = 1.5, and 𝑡 = 3. Look at view at source ↗
Figure 7
Figure 7. Figure 7: The atoms view at source ↗
Figure 10
Figure 10. Figure 10: For more examples that also involve learning the until view at source ↗
Figure 8
Figure 8. Figure 8: Initial dataset of trajectories for learning example. view at source ↗
Figure 9
Figure 9. Figure 9: Learned atoms from the trajectories in Figure 8. view at source ↗
Figure 10
Figure 10. Figure 10: The learned DAG from the trajectories in Figure 8. view at source ↗
Figure 11
Figure 11. Figure 11: Bounds generated by TeLEx for the template in view at source ↗
Figure 13
Figure 13. Figure 13: Showing the boxes used in STL formula generated view at source ↗
Figure 12
Figure 12. Figure 12: Bounds generated by TeLEx for the template in view at source ↗
Figure 14
Figure 14. Figure 14: An example of the DAG simplification algorithm. view at source ↗
Figure 18
Figure 18. Figure 18: The paths are initially around 0; at time view at source ↗
Figure 15
Figure 15. Figure 15: Example 1 trajectories. The 𝑥-axis here represents the time and the 𝑦-axis represents the position in 1D view at source ↗
Figure 20
Figure 20. Figure 20: Example 2 FPL atoms. As expected we can see the learned Atoms captures the diffrent aspects of the trajectory. We also observe that before merging we have multiple atoms corresponding to the final portion of the trajec￾tory where the agent returns to X=0 and stays there for 100 seconds. These atom are merged into a single atom after MergeNodes opera￾tion as shown in Figure 22b. We can visualize the the le… view at source ↗
Figure 21
Figure 21. Figure 21: Example 3 trajectories. (a) The learned DAG for Exam￾ple 3 from the trajectories in view at source ↗
Figure 25
Figure 25. Figure 25: and corresponding atoms as shown in view at source ↗
Figure 23
Figure 23. Figure 23: Learned atoms for Example 3 from the trajectories view at source ↗
Figure 27
Figure 27. Figure 27: Learned DAG for Example 4 after simplification view at source ↗
Figure 28
Figure 28. Figure 28: Learned atoms for Example 4 after simplification view at source ↗
read the original abstract

We introduce a new family of temporal logics intended for specifications in motion planning (MP). It builds upon the signal temporal logic (STL), which is a linear-time logic over real-valued signals that possess quantitative semantics and thus became popular in the areas of cyber-physical systems, robotics, and specifically robot MP. However, in contrast to STL, the proposed logic works with paths as first-class citizens, separating the concerns of geometry and of logic. This in turn leads to simpler and more understandable formulae, and a more refined notion of satisfaction being able to reflect also preferences over behaviours. Technically, the logic is built on fuzzy, time-varying signal constraints. As a consequence of this expressivity, it is (i) more usable for human-given specifications in MP and (ii) more amenable to learning specifications from demonstrations than other logics. The former is important for the traditional style of verification in robot MP; the latter is becoming recognized as crucial for mining data-given tasks and controller synthesis in human-aware MP. We expose the advantages of our proposed logic on examples and show the versatility and flexibility of the framework on a number of scenarios. Finally, we give a learning algorithm with a prototype implementation and discuss the possibilities of model checking and monitoring.

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

0 major / 2 minor

Summary. The manuscript introduces a new family of temporal logics for motion planning that elevates paths to first-class citizens and employs fuzzy time-varying signal constraints, building on but distinguishing from Signal Temporal Logic (STL). The central claims are that this design separates geometric and logical concerns, yielding simpler and more understandable formulae, a satisfaction relation that captures preferences over behaviors, enhanced usability for human-specified requirements, and improved suitability for learning specifications from demonstrations. These advantages are illustrated via examples and scenarios, supported by a learning algorithm and prototype implementation, with discussions on model checking and monitoring.

Significance. Should the proposed logic achieve its stated goals of simplicity and learnability while maintaining a clean separation of concerns, it would offer a meaningful advance in the application of temporal logics to robotics and cyber-physical systems. The provision of concrete examples, a learning procedure with implementation, and exploration of verification directions adds practical value and supports the potential for adoption in motion planning tasks. The constructive presentation, including a prototype, is a strength.

minor comments (2)
  1. [Abstract] The abstract refers to 'a number of scenarios' without specifying how many or their diversity; a brief enumeration or summary table would help readers assess the breadth of the evaluation.
  2. A direct side-by-side comparison of syntax and semantics with STL (perhaps in a dedicated subsection) would make the claimed simplifications more explicit and easier to verify.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their positive summary, recognition of the potential significance for robotics and cyber-physical systems, and recommendation of minor revision. The constructive tone and acknowledgment of the examples, learning procedure, and prototype are appreciated. No major comments were raised in the report.

Circularity Check

0 steps flagged

No significant circularity

full rationale

The paper introduces a new temporal logic by definition, elevating paths to first-class citizens and building it on fuzzy time-varying constraints as an explicit extension of STL. All central claims about simpler formulae, preference-reflecting satisfaction, usability, and learnability are supported directly by the constructive semantics, concrete examples, scenarios, and the supplied learning algorithm. No equations, predictions, or results reduce by construction to fitted parameters or prior self-citations; the framework remains self-contained with temporal and geometric concerns separated by design.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract only; no explicit free parameters, axioms, or invented entities are described. The logic is presented as a new construction on top of STL without detailing any additional postulates.

pith-pipeline@v0.9.0 · 5515 in / 1120 out tokens · 42452 ms · 2026-05-07T17:41:48.405743+00:00 · methodology

discussion (0)

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

Reference graph

Works this paper leans on

62 extracted references · 21 canonical work pages

  1. [1]

    Henzinger

    Rajeev Alur, Tomás Feder, and Thomas A. Henzinger. 1996. The Benefits of Relaxing Punctuality.J. ACM43, 1 (1996), 116–146

    1996

  2. [2]

    Yusuf Aytar, Tobias Pfaff, David Budden, Tom Le Paine, Ziyu Wang, and Nando de Freitas. 2018. Playing hard exploration games by watching YouTube. InPro- ceedings of the 32nd International Conference on Neural Information Processing Systems(Montréal, Canada)(NIPS’18). Curran Associates Inc., Red Hook, NY, USA, 2935–2945

    2018

  3. [3]

    Giuseppe Bombara, Cristian-Ioan Vasile, Francisco Penedo, Hirotoshi Yasuoka, and Calin Belta. 2016. A decision tree approach to data classification using signal temporal logic. InProceedings of the 19th International Conference on Hybrid Systems: Computation and Control. 1–10

    2016

  4. [4]

    Luca Bortolussi, Giuseppe Maria Gallo, Jan Kretínský, and Laura Nenzi. 2022. Learning Model Checking and the Kernel Trick for Signal Temporal Logic on Stochastic Processes. InTACAS (1) (Lecture Notes in Computer Science, Vol. 13243). Springer, 281–300

    2022

  5. [5]

    Ali Tevfik Buyukkocak, Derya Aksaray, and Yasin Yazicioglu. 2021. Control Synthesis using Signal Temporal Logic Specifications with Integral and Deriv- ative Predicates. In2021 American Control Conference, ACC 2021, New Orleans, LA, USA, May 25-28, 2021. IEEE, 4873–4878. https://doi.org/10.23919/ACC50511. 2021.9482651 Logic of Fuzzy Paths

    work page doi:10.23919/acc50511 2021

  6. [6]

    Ali Tevfik Buyukkocak, Derya Aksaray, and Yasin Yazicioglu. 2024. Sequential control barrier functions for mobile robots with dynamic temporal logic spec- ifications.Robotics Auton. Syst.176 (2024), 104681. https://doi.org/10.1016/J. ROBOT.2024.104681

    work page doi:10.1016/j 2024

  7. [7]

    Andreas Bühler, Adrien Gaidon, Andrei Cramariuc, Rares Ambrus, Guy Rosman, and Wolfram Burgard. 2020. Driving Through Ghosts: Behavioral Cloning with False Positives. In2020 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). 5431–5437. https://doi.org/10.1109/IROS45743.2020.9340639

    work page doi:10.1109/iros45743.2020.9340639 2020

  8. [8]

    Cardona, Disha Kamale, and Cristian-Ioan Vasile

    Gustavo A. Cardona, Disha Kamale, and Cristian-Ioan Vasile. 2023. Mixed Integer Linear Programming Approach for Control Synthesis with Weighted Signal Temporal Logic. InProceedings of the 26th ACM International Conference on Hybrid Systems: Computation and Control, HSCC 2023, San Antonio, TX, USA, May 9-12, 2023. ACM, 3:1–3:12. https://doi.org/10.1145/357...

    work page doi:10.1145/3575870.3587120 2023

  9. [9]

    Cardona and Cristian-Ioan Vasile

    Gustavo A. Cardona and Cristian-Ioan Vasile. 2023. Preferences on Partial Satisfaction using Weighted Signal Temporal Logic Specifications. InEuropean Control Conference, ECC 2023, Bucharest, Romania, June 13-16, 2023. IEEE, 1–6. https://doi.org/10.23919/ECC57647.2023.10178201

    work page doi:10.23919/ecc57647.2023.10178201 2023

  10. [10]

    Felipe Codevilla, Matthias Müller, Antonio López, Vladlen Koltun, and Alexey Dosovitskiy. 2018. End-to-End Driving Via Conditional Imitation Learning. In 2018 IEEE International Conference on Robotics and Automation (ICRA). 4693–4700. https://doi.org/10.1109/ICRA.2018.8460487

    work page doi:10.1109/icra.2018.8460487 2018

  11. [11]

    Alexandre Donzé and Oded Maler. 2010. Robust Satisfaction of Temporal Logic over Real-Valued Signals. InFORMATS (Lecture Notes in Computer Science, Vol. 6246). Springer, 92–106

    2010

  12. [12]

    S. Dutta. 1988. An event based fuzzy temporal logic. In[1988] Proceedings. The Eighteenth International Symposium on Multiple-Valued Logic. 64–71. https: //doi.org/10.1109/ISMVL.1988.5151

    work page doi:10.1109/ismvl.1988.5151 1988

  13. [13]

    Georgios Fainekos, Antoine Girard, Hadas Kress-Gazit, and George J. Pappas

  14. [14]

    Temporal logic motion planning for dynamic robots.Autom.45 (2009), 343–352

    2009

  15. [15]

    Fainekos and George J

    Georgios E. Fainekos and George J. Pappas. 2009. Robustness of temporal logic specifications for continuous-time signals.Theor. Comput. Sci.410, 42 (2009), 4262–4291

    2009

  16. [16]

    Achille Frigeri, Liliana Pasquale, and Paola Spoletini. 2014. Fuzzy Time in Linear Temporal Logic.ACM Trans. Comput. Logic15, 4, Article 30 (aug 2014), 22 pages. https://doi.org/10.1145/2629606

    work page doi:10.1145/2629606 2014

  17. [17]

    Barbosa, Jana Tumova, and Jan Kretínský

    Kush Grover, Fernando S. Barbosa, Jana Tumova, and Jan Kretínský. 2021. Se- mantic Abstraction-Guided Motion Planning for scLTL Missions in Unknown Environments. InRobotics: Science and Systems XVII, Virtual Event, July 12- 16, 2021, Dylan A. Shell, Marc Toussaint, and M. Ani Hsieh (Eds.). https: //doi.org/10.15607/RSS.2021.XVII.090

    work page doi:10.15607/rss.2021.xvii.090 2021

  18. [18]

    Kush Grover, Jan Křetínský, Tobias Meggendorfer, and Maximilian Weininger

  19. [19]

    In33rd International Conference on Concurrency Theory (CONCUR

    Anytime Guarantees for Reachability in Uncountable Markov Decision Processes. In33rd International Conference on Concurrency Theory (CONCUR

  20. [20]

    243), Bartek Klin, Sławomir Lasota, and Anca Muscholl (Eds.)

    (Leibniz International Proceedings in Informatics (LIPIcs), Vol. 243), Bartek Klin, Sławomir Lasota, and Anca Muscholl (Eds.). Schloss Dagstuhl – Leibniz- Zentrum für Informatik, Dagstuhl, Germany, 11:1–11:20. https://doi.org/10. 4230/LIPIcs.CONCUR.2022.11

    2022

  21. [21]

    Hans W. Guesgen. 2015. A Fuzzy Set Approach to Expressing Preferences in Spa- tial Reasoning. InAdvances in Knowledge Representation, Logic Programming, and Abstract Argumentation (Lecture Notes in Computer Science, Vol. 9060). Springer, 173–185

    2015

  22. [22]

    Iman Haghighi, Austin Jones, Zhaodan Kong, Ezio Bartocci, Radu Grosu, and Calin Belta. 2015. SpaTeL: a novel spatial-temporal logic and its applications to networked systems. InHSCC. ACM, 189–198

    2015

  23. [23]

    Patrick Halder and Matthias Althoff. 2022. Minimum-Violation Velocity Planning with Temporal Logic Constraints. InITSC. IEEE, 2520–2527

    2022

  24. [24]

    Henzinger, Peter W

    Thomas A. Henzinger, Peter W. Kopke, Anuj Puri, and Pravin Varaiya. 1995. What’s decidable about hybrid automata?. InSTOC. ACM, 373–382

    1995

  25. [25]

    Leibo, and Audrunas Gruslys

    Todd Hester, Matej Vecerik, Olivier Pietquin, Marc Lanctot, Tom Schaul, Bilal Piot, Dan Horgan, John Quan, Andrew Sendonaris, Ian Osband, Gabriel Dulac-Arnold, John Agapiou, Joel Z. Leibo, and Audrunas Gruslys. 2018. Deep Q-learning from demonstrations. InProceedings of the Thirty-Second AAAI Conference on Artificial Intelligence and Thirtieth Innovative ...

    2018

  26. [26]

    Seshia, Tuhin Sahai, and Natarajan Shankar

    Susmit Jha, Ashish Tiwari, Sanjit A. Seshia, Tuhin Sahai, and Natarajan Shankar

  27. [27]

    Hall.Lie Groups, Lie Algebras, and Representations: An Elementary Introduction

    TeLEx: Passive STL Learning Using Only Positive Examples. InRuntime Ver- ification - 17th International Conference, RV 2017, Seattle, W A, USA, September 13- 16, 2017, Proceedings (Lecture Notes in Computer Science, Vol. 10548), Shuvendu K. Lahiri and Giles Reger (Eds.). Springer, 208–224. https://doi.org/10.1007/978-3- 319-67531-2_13

    work page doi:10.1007/978-3- 2017

  28. [28]

    Barbosa, and Jana Tumova

    Jesper Karlsson, Fernando S. Barbosa, and Jana Tumova. 2020. Sampling-based Motion Planning with Temporal Logic Missions and Spatial Preferences.IFAC- PapersOnLine53, 2 (2020), 15537–15543. https://doi.org/10.1016/j.ifacol.2020.12. 2397 21st IFAC World Congress

    work page doi:10.1016/j.ifacol.2020.12 2020

  29. [29]

    Kebria, Abbas Khosravi, Syed Moshfeq Salaken, and Saeid Nahavandi

    Parham M. Kebria, Abbas Khosravi, Syed Moshfeq Salaken, and Saeid Nahavandi

  30. [30]

    https://doi.org/10.1109/JAS.2019.1911825

    Deep imitation learning for autonomous vehicles based on convolutional neural networks.IEEE/CAA Journal of Automatica Sinica7, 1 (2020), 82–95. https://doi.org/10.1109/JAS.2019.1911825

    work page doi:10.1109/jas.2019.1911825 2020

  31. [31]

    Marius Kloetzer and Calin A. Belta. 2008. A Fully Automated Framework for Control of Linear Systems from Temporal Logic Specifications.IEEE Trans. Automat. Control53 (2008), 287–297

    2008

  32. [32]

    Zhaodan Kong, Austin Jones, Ana Medina Ayala, Ebru Aydin Gol, and Calin Belta. 2014. Temporal logic inference for classification and prediction from data. InProceedings of the 17th international conference on Hybrid systems: computation and control. 273–282

    2014

  33. [33]

    Hadas Kress-Gazit, Georgios Fainekos, and George J. Pappas. 2009. Temporal- Logic-Based Reactive Mission and Motion Planning.IEEE Transactions on Robotics 25 (2009), 1370–1381

    2009

  34. [34]

    1991.Robot motion planning

    Jean-Claude Latombe. 1991.Robot motion planning

    1991

  35. [35]

    Steven M. LaValle. 2006.Planning algorithms

    2006

  36. [36]

    Kevin Leahy, Eric Cristofalo, Cristian Ioan Vasile, Austin Jones, Eduardo Monti- jano, Mac Schwager, and Calin Belta. 2019. Control in belief space with temporal logic specifications using vision-based localization.Int. J. Robotics Res.38, 6 (2019). https://doi.org/10.1177/0278364919846340

    work page doi:10.1177/0278364919846340 2019

  37. [37]

    Yongming Li and Jielin Wei. 2021. Possibilistic Fuzzy Linear Temporal Logic and Its Model Checking.IEEE Transactions on Fuzzy Systems29, 7 (2021), 1899–1913. https://doi.org/10.1109/TFUZZ.2020.2988848

    work page doi:10.1109/tfuzz.2020.2988848 2021

  38. [38]

    Alexis Linard, Ilaria Torre, Iolanda Leite, and Jana Tumova. 2022. Inference of Multi-Class STL Specifications for Multi-Label Human-Robot Encounters. In IROS. IEEE, 1305–1311

    2022

  39. [39]

    Dimarogonas

    Lars Lindemann and Dimos V. Dimarogonas. 2019. Control Barrier Functions for Signal Temporal Logic Tasks.IEEE Control Systems Letters3 (2019), 96–101

    2019

  40. [40]

    Dimarogonas

    Lars Lindemann and Dimos V. Dimarogonas. 2019. Robust control for signal temporal logic specifications using discrete average space robustness.Autom. 101 (2019), 377–387

    2019

  41. [41]

    Rudolf Lioutikov, Gerhard Neumann, Guilherme Maeda, and Jan Peters. 2017. Learning movement primitive libraries through probabilistic segmentation.The International Journal of Robotics Research36, 8 (2017), 879–894. https://doi.org/ 10.1177/0278364917713116 arXiv:https://doi.org/10.1177/0278364917713116

    work page doi:10.1177/0278364917713116 2017

  42. [42]

    Oded Maler and Dejan Nickovic. 2004. Monitoring Temporal Properties of Continuous Signals. InFORMATS/FTRTFT (Lecture Notes in Computer Science, Vol. 3253). Springer, 152–166

    2004

  43. [43]

    Noushin Mehdipour, Cristian-Ioan Vasile, and Calin Belta. 2019. Average-based robustness for continuous-time signal temporal logic. In2019 IEEE 58th Confer- ence on Decision and Control (CDC). IEEE, 5312–5317

    2019

  44. [44]

    Noushin Mehdipour, Cristian-Ioan Vasile, and Calin Belta. 2021. Specifying User Preferences Using Weighted Signal Temporal Logic.IEEE Control Systems Letters 5, 6 (2021), 2006–2011. https://doi.org/10.1109/LCSYS.2020.3047362

    work page doi:10.1109/lcsys.2020.3047362 2021

  45. [45]

    Takayuki Osa, Naohiko Sugita, and Mamoru Mitsuishi. 2014. Online Trajectory Planning in Dynamic Environments for Surgical Task Automation.. InRobotics: Science and Systems. 1–9

    2014

  46. [46]

    Amir Pnueli. 1977. The Temporal Logic of Programs. In18th Annual Symposium on Foundations of Computer Science, Providence, Rhode Island, USA, 31 October - 1 November 1977. IEEE Computer Society, 46–57. https://doi.org/10.1109/SFCS. 1977.32

    work page doi:10.1109/sfcs 1977

  47. [47]

    Murray, Alberto L

    Vasumathi Raman, Alexandre Donzé, Mehdi Maasoumy, Richard M. Murray, Alberto L. Sangiovanni-Vincentelli, and Sanjit A. Seshia. 2014. Model predictive control with signal temporal logic specifications.53rd IEEE Conference on Decision and Control(2014), 81–87

    2014

  48. [48]

    Dorsa Sadigh and Ashish Kapoor. 2016. Safe control under uncertainty with probabilistic signal temporal logic. InProceedings of Robotics: Science and Systems XII

    2016

  49. [49]

    Guy Scher, Sadra Sadraddini, and Hadas Kress-Gazit. 2022. Robustness-based Synthesis for Stochastic Systems under Signal Temporal Logic Tasks. InIROS. IEEE, 1269–1275

    2022

  50. [50]

    Henning Schlüter, Philipp Schillinger, and Mathias Bürger. 2018. On the Design of Penalty Structures for Minimum-Violation LTL Motion Planning. InCDC. IEEE, 4153–4158

    2018

  51. [51]

    Ho Chit Siu, Kevin Leahy, and Makai Mann. 2023. STL: Surprisingly Tricky Logic (for System Validation). InIROS. 8613–8620. https://doi.org/10.1109/IROS55552. 2023.10342290

    work page doi:10.1109/iros55552 2023

  52. [52]

    Smith, Jana Tumova, Calin A

    Stephen L. Smith, Jana Tumova, Calin A. Belta, and Daniela Rus. 2010. Optimal path planning under temporal logic constraints.2010 IEEE/RSJ International Conference on Intelligent Robots and Systems(2010), 3288–3293

    2010

  53. [53]

    Sadegh Esmaeil Zadeh Soudjani and Alessandro Abate. 2011. Adaptive Gridding for Abstraction and Verification of Stochastic Hybrid Systems. In2011 Eighth International Conference on Quantitative Evaluation of SysTems. 59–68. https: //doi.org/10.1109/QEST.2011.16

    work page doi:10.1109/qest.2011.16 2011

  54. [54]

    Gordon, Byron Boots, and J

    Wen Sun, Arun Venkatraman, Geoffrey J. Gordon, Byron Boots, and J. Andrew Bagnell. 2017. Deeply AggreVaTeD: Differentiable Imitation Learning for Sequen- tial Prediction. InProceedings of the 34th International Conference on Machine Learning (Proceedings of Machine Learning Research, Vol. 70), Doina Precup and Kush Grover, Pratham Gupta, and Jan Křetínský...

    2017

  55. [55]

    Reyes Castro, Sertac Karaman, Emilio Frazzoli, and Daniela Rus

    Jana Tumova, Luis I. Reyes Castro, Sertac Karaman, Emilio Frazzoli, and Daniela Rus. 2013. Minimum-violation LTL planning with conflicting specifications. In ACC. IEEE, 200–205

    2013

  56. [56]

    DeLa- teur, Ron Weiss, Douglas Densmore, and Calin Belta

    Prashant Vaidyanathan, Rachael Ivison, Giuseppe Bombara, Nicholas A. DeLa- teur, Ron Weiss, Douglas Densmore, and Calin Belta. 2017. Grid-based temporal logic inference. In2017 IEEE 56th Annual Conference on Decision and Control (CDC). 5354–5359. https://doi.org/10.1109/CDC.2017.8264452

    work page doi:10.1109/cdc.2017.8264452 2017

  57. [57]

    Cristian Ioan Vasile, Vasumathi Raman, and Sertac Karaman. 2017. Sampling- based synthesis of maximally-satisfying controllers for temporal logic specifica- tions. InIROS. IEEE, 3840–3847

    2017

  58. [58]

    Cristian Ioan Vasile, Jana Tumova, Sertac Karaman, Calin Belta, and Daniela Rus

  59. [59]

    Minimum-violation scLTL motion planning for mobility-on-demand. In ICRA. IEEE, 1481–1488

  60. [60]

    Tichakorn Wongpiromsarn, Konstantin Slutsky, Emilio Frazzoli, and Ufuk Topcu

  61. [61]

    Minimum-Violation Planning for Autonomous Systems: Theoretical and Practical Considerations. InACC. IEEE, 4866–4872

  62. [62]

    independent vari- ables

    Boyuan Zheng, Sunny Verma, Jianlong Zhou, Ivor W. Tsang, and Fang Chen. 2022. Imitation Learning: Progress, Taxonomies and Challenges.IEEE Transactions on Neural Networks and Learning Systems(2022), 1–16. https://doi.org/10.1109/ TNNLS.2022.3213246 A Semantics A.1 Computing Exact Semantics We discuss the integration and maximum semantics separately here s...

    work page arXiv 2022