'Black technology' appears at the Greater Bay Area Rail Transit Forum, smart transportation moves towards 'strong AI'

Exhibition site Photo by Liao Yunna

China News Service, Guangzhou, June 29 (Wang Jian and Cen Baixian) "Black technology" covering urban rail vehicles, signaling systems, smart energy, smart stations and other transportation technology fields will be unveiled at the Guangzhou Greater Bay Area Rail Transit Forum and Conference on the 29th The exhibition brought together nearly 300 exhibitors including Guangzhou Metro, Shanghai STO, CRRC, Jiadu Technology, and Baiyun Electric, with a total exhibition area of ​​over 20,000 square meters.

At the exhibition site, advanced technologies such as 80 kilometers per hour Type A Chinese standard subway train, AI-based BIM intelligent review platform, rail transit integrated intelligent power monitoring system, urban rail transit power supply two-way AC device, and urban rail transit platform door system Many viewers applauded and marveled.

The technology that attracted the most attention from the audience was undoubtedly the outstanding capabilities of "strong artificial intelligence" demonstrated in the field of transportation, which attracted a large number of viewers to stop and watch. Whether it is real-time information Q&A in passenger service scenarios, overall planning of emergency command scenarios, or complex fault handling tasks in subway operation and maintenance scenarios, the Jiadu Zhixing large model (hereinafter referred to as the "big model") was released for the first time on this forum. All "double business" is online.

Jiadu Technology stated that the capabilities of large-scale intelligent models have surpassed people's traditional understanding of artificial intelligence and are gradually moving towards the stage of "strong artificial intelligence". The large model adopts hybrid cloud multi-mode technology to form a three-layer architecture of "large model infrastructure", "large model middleware" and "large model industry application" as a whole, thereby achieving reinforcement learning of "pre-training and fine-tuning based on industry data feedback".

At the same time, this model currently has mature application capabilities in three scenarios: intelligent customer service, intelligent operation and maintenance, and emergency command for rail transit.

The reporter learned at the scene that this model can be transformed into a "digital operation and maintenance expert" to talk to maintenance personnel, thereby assisting them in troubleshooting the cause of faults, analyzing solutions and providing auxiliary maintenance decisions; it can also build a large language model to support auxiliary capabilities. It can provide urban rail emergency response services before, during and after the incident. (over)

The above is the detailed content of 'Black technology' appears at the Greater Bay Area Rail Transit Forum, smart transportation moves towards 'strong AI'. For more information, please follow other related articles on the PHP Chinese website!