TOSHKENT SHAHRI TRANSPORT TIZIMINI OPTIMIZALLASHTIRISH UCHUN GIBRID INTELLEKTUAL MODEL: BASHORATLI TAHLIL VA CHUQUR MUSTAHKAMLOVCHI TA'LIM INTEGRATSIYASI
DOI:
https://doi.org/10.65164/vngv7c29Keywords:
aqlli transport tizimlari, transport oqimini optimallashtirish, chuqur mustahkamlovchi ta'lim, LSTM, ko‘p agentli tizimlar, shahar simulyatsiyasi, SUMO, Toshkent.Abstract
Katta shaharlardagi transport tirbandligi zamonaviy urbanizatsiyaning eng dolzarb muammolaridan biri bo‘lib, iqtisodiy yo‘qotishlar va ekologik zararga olib keladi. Mavjud transport oqimini boshqarish tizimlari, asosan, qat'iy belgilangan vaqt sikllariga yoki lokal datchiklardan olingan real vaqt ma'lumotlariga tayanadi, bu esa ularning dinamik va oldindan aytib bo‘lmaydigan transport sharoitlariga moslashish qobiliyatini cheklaydi. Ushbu tadqiqotda Toshkent shahri misolida transport oqimini proaktiv boshqarish uchun LSTM (Long Short-Term Memory) va MADRL (Multi-Agent Deep Reinforcement Learning) arxitekturalarini birlashtirgan gibrid intellektual model taklif etiladi. Modelning bashorat komponenti yaqin kelajakdagi transport oqimining zichligini bashorat qilish uchun LSTM tarmog'idan foydalanadi. Boshqaruv komponenti esa har bir chorrahani mustaqil agent sifatida modellashtiruvchi MADRL tizimidan iborat bo‘lib, u bashorat qilingan ma'lumotlar va qo‘shni agentlarning holatini hisobga olgan holda svetofor fazalarini optimal boshqarishga o‘rganadi. Taklif etilayotgan modelning samaradorligi SUMO (Simulation of Urban MObility) platformasida yaratilgan Toshkent shahrining markaziy qismi transport tarmog'i simulyatsiyasi yordamida baholanadi. Dastlabki natijalar shuni ko‘rsatadiki, gibrid model an'anaviy qat'iy vaqtli va reaktiv tizimlarga qaraganda transport vositalarining o‘rtacha kutish vaqtini va yo‘ldagi umumiy vaqtni sezilarli darajada kamaytirish imkoniyatiga ega.
References
T.A. Хужакулов, Р.Т.Гаипназаров “Большие данные (Big Data) и искусственный интеллект
(AI) для интеллектуального управления городским транспортом в Узбекистане”, Digital
transformation: a new era in information technology, artificial intelligence and the economy.
International scientific-practical conference, pp. 430-438, 2025
2. Постановление Президента Республики Узбекистан. "О внедрении системы цифрового
управления дорожным движением в г.Ташкенте" (В Ташкенте внедрят цифровую систему
управления дорожным движением) (проект, 2023).
3. S. S. Abdulhai, B. "Reinforcement learning for true adaptive traffic signal control," Journal of
Transportation Engineering, vol. 129, no. 3, pp. 278-285, 2003.
4. E. van der Pol, F. A. Oliehoek, "Coordinated deep reinforcement learners for traffic light control,"
Proceedings of the 15th International Conference on Autonomous Agents and Multiagent
Systems (AAMAS), 2016.
5. B. Williams, M. "Modeling and forecasting vehicular traffic flow as a seasonal ARIMA process:
Theoretical basis and empirical results," Journal of Transportation Engineering, vol. 129, no. 6,
pp. 664-672, 2003.
6. Y. Tian, Z. "LSTM-based traffic flow prediction with missing data," Neurocomputing, vol. 318,
pp. 297-305, 2018.
7. P. A. Lopez et al., "Microscopic Traffic Simulation using SUMO," Proceedings of the 21st IEEE
International Conference on Intelligent Transportation Systems (ITSC), 2018.
8. S. Hochreiter, J. Schmidhuber, "Long short-term memory," Neural Computation, vol. 9, no. 8,
pp. 1735-1780, 1997.
9. R. S. Sutton, A. G. Barto, Reinforcement Learning: An Introduction, 2nd ed. MIT Press, 2018.
10. D. Schrank, B. Eisele, T. Lomax, "2019 Urban Mobility Report," Texas A&M Transportation
Institute, 2019.
11. L. G. Gazis, D. C. "The control of traffic flow in tunnels," Transportation Science, vol. 3, no. 1,
pp. 48-64, 1969.
