Abstract
This paper describes a formal model of a “location, heading and speed” waypoint navigation task for an autonomous ground vehicle—that is, a task of navigating the vehicle towards a particular location so that it has the desired heading and speed when in that location. Our novel way of modeling this task makes formal reasoning over controller correctness tractable. We state our model in differential dynamic logic (dL), which we then use to establish a formal definition of waypoint feasibility and formally verify its validity in the KeYmaera X interactive theorem prover. The formal machine-checked proof witnesses that for any waypoint we consider feasible, the vehicle can indeed be controlled to reach it within the prescribed error bound. We also describe how we use these formal definitions and theorem statements to inform training of neural network controllers for performing this waypoint navigation task. Note that in our approach we do not need to rely on the neural network controller always being perfect—instead, the formal model allows a synthesis of a correct-by-construction safety net for the controller that checks whether the neural network output is safe to act upon and present a safe alternative if it is not.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
As a consequence of instantaneous steering, the curvature of the vehicle’s path is able to change instantaneously to any value in the feasible range (that is, between \(-\frac{1}{R_\texttt {min}}\) and \(\frac{1}{R_\texttt {min}}\), where \(R_\texttt {min}\) is the minimum turning radius of the vehicle).
- 2.
Some solvers, e.g. dReal [14], opt for \(\delta \)-decidability to render transcendental functions decidable.
References
Achiam, J., Held, D., Tamar, A., Abbeel, P.: Constrained policy optimization. In: Proceedings of the 34th International Conference on Machine Learning, ICML 2017, Sydney, NSW, Australia, pp. 22–31 (2017). proceedings.mlr.press/v70/achiam17a.html
Ahn, E.: Towards Safe Reinforcement Learning in the Real World. Master’s thesis, Carnegie Mellon University, cMU-RI-TR-19-56 (2019). www.ri.cmu.edu/wp-content/uploads/2019/08/MSR/Thesis/Edward/Ahn/2019.pdf
Alonso-Mora, J., DeCastro, J.A., Raman, V., Rus, D., Kress-Gazit, H.: Reactive mission and motion planning with deadlock resolution avoiding dynamic obstacles. Auton. Robot. 42(4), 801–824 (2017). https://doi.org/10.1007/s10514-017-9665-6
Bohrer, B., Rahli, V., Vukotic, I., Völp, M., Platzer, A.: Formally verified differential dynamic logic. In: Proceedings of the 6th ACM SIGPLAN Conference on Certified Programs and Proofs. ACM (2017). https://doi.org/10.1145/3018610.3018616
Bohrer, B., Tan, Y.K., Mitsch, S., Sogokon, A., Platzer, A.: A formal safety net for waypoint following in ground robots. IEEE Robot. Automat. Lett. 4(3), 2910–2917 (2019). https://doi.org/10.1109/LRA.2019.2923099
Chang, Y.C., Gao, S.: Stabilizing neural control using self-learned almost Lyapunov critics. In: Proceedings of the 2021 International Conference on Robotics and Automation (ICRA 2021) (2021). arxiv.org/abs/2107.04989
Chang, Y.C., Roohi, N., Gao, S.: Neural Lyapunov control (2020). arxiv.org/abs/2005.00611
Davenport, J.H., Heintz, J.: Real quantifier elimination is doubly exponential. J. Symb. Comput. 5(1/2), 29–35 (1988). https://doi.org/10.1016/S0747-7171(88)80004-X
Desai, A., Saha, I., Yang, J., Qadeer, S., Seshia, S.A.: DRONA: a framework for safe distributed mobile robotics. In: Martínez, S., Tovar, E., Gill, C., Sinopoli, B. (eds.) Proceedings of the 8th International Conference on Cyber-Physical Systems, ICCPS 2017, Pittsburgh, Pennsylvania, USA, pp. 239–248. ACM (2017). https://doi.org/10.1145/3055004.3055022
Dreossi, T., Donzé, A., Seshia, S.A.: Compositional falsification of cyber-physical systems with machine learning Components. J. Autom. Reason. 63(4), 1031–1053 (2019). https://doi.org/10.1007/s10817-018-09509-5
Fisher, M., Mascardi, V., Rozier, K.Y., Schlingloff, B.-H., Winikoff, M., Yorke-Smith, N.: Towards a framework for certification of reliable autonomous systems. Auton. Agents Multi-Agent Syst. 35(1), 1–65 (2020). https://doi.org/10.1007/s10458-020-09487-2
Foughali, M., Bensalem, S., Combaz, J., Ingrand, F.: Runtime verification of timed properties in autonomous robots. In: 18th ACM/IEEE International Conference on Formal Methods and Models for System Design, MEMOCODE 2020, Jaipur, India, pp. 1–12. IEEE (2020). https://doi.org/10.1109/MEMOCODE51338.2020.9315156
Fulton, N., Mitsch, S., Quesel, J., Völp, M., Platzer, A.: Keymaera X: an axiomatic tactical theorem prover for hybrid systems. In: Felty, A.P., Middeldorp, A. (eds.) Automated Deduction - CADE-25 - 25th International Conference on Automated Deduction, Berlin, Germany, 2015, Proceedings. Lecture Notes in Computer Science, vol. 9195, pp. 527–538. Springer (2015). https://doi.org/10.1007/978-3-319-21401-6_36
Gao, S., Kong, S., Clarke, E.M.: dreal: An SMT solver for nonlinear theories over the reals. In: Bonacina, M.P. (ed.) Automated Deduction - CADE-24 - 24th International Conference on Automated Deduction, Lake Placid, NY, USA, 2013. Proceedings. Lecture Notes in Computer Science, vol. 7898, pp. 208–214. Springer (2013). https://doi.org/10.1007/978-3-642-38574-2_14
Hoxha, B., Fainekos, G.E.: Planning in dynamic environments through temporal logic monitoring. In: Magazzeni, D., Sanner, S., Thiébaux, S. (eds.) Planning for Hybrid Systems, Papers from the 2016 AAAI Workshop, Phoenix, Arizona, USA 2016. AAAI Workshops, vol. WS-16-12. AAAI Press (2016). www.aaai.org/ocs/index.php/WS/AAAIW16/paper/view/12556
Huang, J., et al.: ROSRV: runtime verification for robots. In: Bonakdarpour, B., Smolka, S.A. (eds.) RV 2014. LNCS, vol. 8734, pp. 247–254. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-11164-3_20
Ivanov, R., Carpenter, T.J., Weimer, J., Alur, R., Pappas, G.J., Lee, I.: Verifying the safety of autonomous systems with neural network controllers. ACM Trans. Embed. Comput. Syst. 20(1), 1–26 (2021). https://doi.org/10.1145/3419742
Kania, R., Frederick, P., Pritchett, W., Wood, B., Mentzer, C., Johnson, E.: Dismounted soldier autonomy tools (DSAT) – from conception to deployment. In: 2014 NDIA Ground Vehicles Systems Engineering and Technology Symposium (2014). gvsets.ndia-mich.org/publication.php?documentID=171
Kempa, B., Zhang, P., Jones, P.H., Zambreno, J., Rozier, K.Y.: Embedding online runtime verification for fault disambiguation on robonaut2. In: Bertrand, N., Jansen, N. (eds.) Formal Modeling and Analysis of Timed Systems - 18th International Conference, FORMATS 2020, Vienna, Austria, Proceedings. Lecture Notes in Computer Science, vol. 12288, pp. 196–214. Springer (2020). https://doi.org/10.1007/978-3-030-57628-8_12
Liu, S.B., Roehm, H., Heinzemann, C., Lütkebohle, I., Oehlerking, J., Althoff, M.: Provably safe motion of mobile robots in human environments. In: 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2017, Vancouver, BC, Canada, pp. 1351–1357. IEEE (2017). https://doi.org/10.1109/IROS.2017.8202313
Luckcuck, M., Farrell, M., Dennis, L.A., Dixon, C., Fisher, M.: Formal specification and verification of autonomous robotic systems: a survey. ACM Comput. Surv. 52(5), 1–41 (2019). https://doi.org/10.1145/3342355
Milner, R.: LCF: a way of doing proofs with a machine. In: Becvár, J. (ed.) Mathematical Foundations of Computer Science 1979, Proceedings, 8th Symposium Lecture Notes in Computer Science. Olomouc, Czechoslovakia, vol. 74, pp. 146–159. Springer (1979). https://doi.org/10.1007/3-540-09526-8_11
Mitsch, S., Ghorbal, K., Vogelbacher, D., Platzer, A.: Formal verification of obstacle avoidance and navigation of ground robots. I. J. Robotics Res. 36(12), 1312–1340 (2017). https://doi.org/10.1177/0278364917733549
Mitsch, S., Platzer, A.: ModelPlex: verified runtime validation of verified cyber-physical system models. Formal Methods Syst. Design (1), 33–74 (2016). https://doi.org/10.1007/s10703-016-0241-z
Pan, Y., Lin, Q., Shah, H., Dolan, J.M.: Safe planning for self-driving via adaptive constrained ILQR. CoRR abs/2003.02757 (2020). arxiv.org/abs/2003.02757
Platzer, A.: Differential dynamic logic for hybrid systems. J. Autom. Reasoning 41(2), 143–189 (2008). https://doi.org/10.1007/s10817-008-9103-8
Platzer, A.: A Complete Uniform Substitution Calculus for Differential Dynamic Logic. J. Autom. Reason. 59(2), 219–265 (2016). https://doi.org/10.1007/s10817-016-9385-1
Platzer, A.: Logical Foundations of Cyber-Physical Systems. Springer (2018). https://doi.org/10.1007/978-3-319-63588-0
Platzer, A., Clarke, E.M.: Formal verification of curved flight collision avoidance maneuvers: a case study. In: Cavalcanti, A., Dams, D. (eds.) FM. LNCS, vol. 5850, pp. 547–562. Springer (2009). https://doi.org/10.1007/978-3-642-05089-3_35
Platzer, A., Tan, Y.K.: Differential equation invariance axiomatization. J. ACM 67(1), 1–66 (2020). https://doi.org/10.1145/3380825
Rizaldi, A., Immler, F., Schürmann, B., Althoff, M.: A formally verified motion planner for autonomous vehicles. In: Lahiri, S.K., Wang, C. (eds.) Automated Technology for Verification and Analysis - 16th International Symposium, ATVA 2018, Los Angeles, CA, USA, 2018, Proceedings. Lecture Notes in Computer Science, vol. 11138, pp. 75–90. Springer (2018). https://doi.org/10.1007/978-3-030-01090-4_5
Schulman, J., Levine, S., Moritz, P., Jordan, M.I., Abbeel, P.: Trust region policy optimization. CoRR abs/1502.05477 (2015). arxiv.org/abs/1502.05477
Schulman, J., Wolski, F., Dhariwal, P., Radford, A., Klimov, O.: Proximal policy optimization algorithms (2017). arxiv.org/abs/1707.06347v2
Seegmiller, N.: Dynamic model formulation and calibration for wheeled mobile robots. Ph.D. thesis, Carnegie Mellon University, Pittsburgh, PA (2014). www.ri.cmu.edu/publications/dynamic-model-formulation-and-calibration-for-wheeled-mobile-robots/
Seegmiller, N., Kelly, A.: High-fidelity yet fast dynamic models of wheeled mobile robots. IEEE Trans. Robot. 32(3), 614–625 (2016). https://doi.org/10.1109/TRO.2016.2546310
Shivakumar, S., Torfah, H., Desai, A., Seshia, S.A.: SOTER on ROS: a run-time assurance framework on the robot operating system. In: Deshmukh, J., Ničković, D. (eds.) RV 2020. LNCS, vol. 12399, pp. 184–194. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-60508-7_10
Tran, H., et al.: NNV: the neural network verification tool for deep neural networks and learning-enabled cyber-physical systems. In: Lahiri, S.K., Wang, C. (eds.) Computer Aided Verification - 32nd International Conference, CAV 2020, Los Angeles, CA, USA, 2020, Proceedings, Part I. Lecture Notes in Computer Science, vol. 12224, pp. 3–17. Springer (2020). https://doi.org/10.1007/978-3-030-53288-8_1
Tuncali, C.E., Fainekos, G., Prokhorov, D.V., Ito, H., Kapinski, J.: Requirements-driven test generation for autonomous vehicles with machine learning components. IEEE Trans. Intell. Veh. 5(2), 265–280 (2020). https://doi.org/10.1109/TIV.2019.2955903
Tuncali, C.E., Kapinski, J., Ito, H., Deshmukh, J.V.: Reasoning about safety of learning-enabled components in autonomous cyber-physical systems. In: Proceedings of the 55th Annual Design Automation Conference, DAC 2018, San Francisco, CA, USA, pp. 1–6. ACM (2018). https://doi.org/10.1145/3195970.3199852
Wong, K.W., Ehlers, R., Kress-Gazit, H.: Resilient, provably-correct, and high-level robot behaviors. IEEE Trans. Robot. 34(4), 936–952 (2018). https://doi.org/10.1109/TRO.2018.2830353
Wong, K.W., Finucane, C., Kress-Gazit, H.: Provably-correct robot control with ltlmop, OMPL and ROS. In: 2013 IEEE/RSJ International Conference on Intelligent Robots and Systems, Tokyo, Japan, p. 2073. IEEE (2013). https://doi.org/10.1109/IROS.2013.6696636
Acknowledgment
This material is based upon work supported by the United States Air Force and DARPA under Contract No. FA8750-18-C-0092. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the view of the United States Air Force and DARPA. Distribution Statement “A” (Approved for Public Release, Distribution Unlimited).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Kopylov, A., Mitsch, S., Nogin, A., Warren, M. (2021). Formally Verified Safety Net for Waypoint Navigation Neural Network Controllers. In: Huisman, M., Păsăreanu, C., Zhan, N. (eds) Formal Methods. FM 2021. Lecture Notes in Computer Science(), vol 13047. Springer, Cham. https://doi.org/10.1007/978-3-030-90870-6_7
Download citation
DOI: https://doi.org/10.1007/978-3-030-90870-6_7
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-90869-0
Online ISBN: 978-3-030-90870-6
eBook Packages: Computer ScienceComputer Science (R0)