The article has considered the approach to the guaranteed vessels collision avoidance problem on the basis of possible collisions sets in speed space. The control task is to safely maneuver the vessel in a navigational envi-ronment with many potentially dangerous target vessels. The area of possible collision is the area of the physical water space in which a collision with a given target vessel is possible, as a function of time. The controlled vessel must avoid any set of possible collisions at any given time. The union of all sets of possible collisions at each moment of time forms a two-dimensional set in the speed space, which we will call an obstacle in the speed space. If the controlled vessel is moving at a constant heading and speed, then for guaranteed collision avoidance, the speed vector should not be inside the obstacle area in speed space at any time from a given planning horizon. Collision avoidance at large planning horizons can be ensured through the use of permanent safe states, which guarantee the safety of navigation and are used in iterative planning algorithms. To manage this development, many iterative motion planners define constant safe states as sets of controlled ship states for which safety (collision avoidance) conditions are satisfied, and transitions back to these states are dynamically feasible. Moreover, for guaranteed collision avoidance, the concept of an imminent collision state is used, which is closely related to constant safe states. In general, the safety of navigation is achieved by defining a space of safe states and restrictions on control, allowing the ship to be only within this space. Sequential estimation of navigation environment and iterative planning algorithms guarantee safety for infinite planning horizon. Rec-ommendations for practical use have been given. Performed researches contribute to the improvement of the ship’s handling methods
speed space, collision avoidance, ship handling, safe states
1. Zhuk A.S. Model' dvizheniya sudna po pro-gramme vyhoda v putevuyu tochku / A.S. Zhuk // Transport: nauka, tehnika, upravlenie. Nauchnyy informacionnyy sbornik.- 2020. - № 6. - S. 32-36.
2. Tolpegin O.A. Upravlenie raketami na os-nove rascheta oblastey dostizhimosti / O.A. Tolpegin // Vestnik Samarskogo gosudar-stvennogo aerokosmicheskogo universi-teta.- 2015. - T.14.- № 1.- S. 73-82.
3. Damas B. Avoiding moving obstacles: the for-bidden velocity map / B. Damas, J. Santos-Victor // IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE. - 2009. - P. 4393-4398.
4. Fiorini P. Motion planning in dynamic envi-ronments using the relative velocity paradigm / P. Fiorini, Z. Shiller // IEEE International Conference on Robotics and Automation (ICRA). IEEE. - 1993. - P. 560-565.
5. Fiorini P. Motion planning in dynamic envi-ronments using velocity obstacles / P. Fiorini, Z. Shiller // The International Journal of Ro-botics Research. - 1998. - vol.17, No. 7. - P. 760.
6. Gal O. Efficient and safe on-line motion plan-ning in dynamic environments / O. Gal, Z. Shiller, E. Rimon // IEEE International Con-ference on Robotics and Automation. (ICRA) IEEE. - 2009. - P. 88 - 93.
7. Large F. Navigation among moving obstacles using the NLVO: Principles and applications to intelligent vehicles / F. Large, C. Laugier, Z. Shiller // Autonomous Robots. - 2005. - vol. 19, No. 2. - P. 159 - 171.
8. Wu A. Guaranteed Avoidance of Unpredicta-ble, Dynamically Constrained Obstacles using Velocity Obstacle Sets / A. Wu. - Massachu-setts: Massachusetts Institute of Technology, 2011.-116p.
9. Zhuk A.S. Model' trehmernogo mnozhestva dostizhimosti dvizheniya sudna / A.S. Zhuk // Ekspluataciya morskogo transporta. - 2017. - № 2 (83). - S. 51-57.
10. Zhuk A.S. Trehmernye mnozhestva dostizhi-mosti i prityazheniya sudna s ogranichennym upravleniem / A.S. Zhuk // Ekspluataciya morskogo transporta. - 2018. - № 2 (87). - S. 32-38.
11. Holmes P. Reachable Sets for Safe, Real-Time Manipulator Trajectory Design / P.Holmes, S.Kousik, B.Zhang, D.Raz, C.Barbalata, M.Johnson-Roberson, R.Vasudevan // Robot-ics: Science and Systems, 2020.
12. Allen R.E. A Machine Learning Approach for Real-Time Reachability Analysis / R.E.Allen, A.A.Clark, J.A.Starek, M.Pavone // 2014 IEEE/RSJ International Conference on Intelli-gent Robots and Systems. IEEE. 2014.
13. Hua C. Reachable set modeling and engage-ment analysis of exoatmospheric interceptor / C.Hua, L.Yangang, C.Lei, T.Guojin // Chinese Journal of Aeronautics.- 2014. 27(6). P. 1513 - 1526.
14. Lin Y. Collision avoidance for UAVs using reachable sets / Y.Lin, S.Saripalli // 2015 In-ternational Conference on Unmanned Aircraft Systems (ICUAS). IEEE. 2015.
15. Boran-Keshish'yan A.L. Analiz modeley upravleniya potokom sudov v morskih por-tah v razvitii koncepcii e-navigacii / A.L. Boran-Keshish'yan, V.V. Popov // Eks-pluataciya morskogo transporta.- 2018.- № 1 (86).- S. 91-95.
16. Fraichard T. Inevitable collision states - a step towards safer robots? / T. Fraichard, H. Asama // Advanced Robotics. - 2004. - vol. 18, No. 10. - P. 1001 - 1024.
17. Petti S. Safe motion planning in dynamic en-vironments / S. Petti, T. Fraichard // IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE. - 2005. - P. 2210 - 2215.
18. Astrein V.V. Multicriteria assessment of opti-mal forecasting models in decision support systems to ensure the navigation safety / V.V. Astrein, S.I. Kondratyev, A.L. Boran-Keshishyan // Journal of Physics: Conference Series. "International Conference on Actual Issues of Mechanical Engineering, AIME 2021". - 2021. - P. 012108.
19. Dantsevich I. Numerical Method for Correct-ing Command Signals for Combined Control of a Multiengined Complex / I. Dantsevich, M. Lyutikova, V. Fedorenko // International Con-ference on Mathematics and its Applications in new Computer Systems. Springer, Cham. - 2022. - P. 117-131.
20. Kondratyev S.I. Human-machine system as a control shell in the implementation of mooring operations / S.I. Kondratyev, A.L. Boran-Keshishyan, V.V. Popov, A.E. Slitsan // Jour-nal of Physics: Conference Series. "Interna-tional Conference on Actual Issues of Me-chanical Engineering, AIME 2021". - 2021. - P. 012045
