en.Wedoany.com Reported - China's Bronto Technology Co., Ltd. (hereinafter referred to as "Bronto") recently unveiled a universal autonomous driving AI model for mines. Chen Fangming, the company's chairman, explained the underlying logic and core breakthroughs of this technological innovation in an exclusive interview with Securities Daily, outlining a new path for mine intelligence characterized by "born unmanned, beyond unmanned." This approach aims to address the industry pain point where traditional single-vehicle automation solutions fail to adapt to the complex and dynamic production demands of mining sites.
Mines represent a typical complex and dynamic operational scenario in the industrial sector. Continuous advancement of mining faces leads to evolving road layouts, daily updates to loading and dumping areas, and intricate interactions between mining trucks, excavators, auxiliary equipment, and personnel. Traditional autonomous driving technology for mines focuses solely on solving the basic problem of "vehicle self-driving," relying on fixed programs and preset rules. This results in significant shortcomings in adaptability and stability when faced with a vast number of long-tail scenarios, sudden operating conditions, and complex environments in mines.
The universal autonomous driving AI model released by Bronto at the end of June reconstructs the intelligent core of unmanned mining trucks. Based on a new intelligent system, these trucks can precisely adapt to production needs across different operational scenarios. For example, in loading areas, they can autonomously assess the rhythm of excavator operations and accurately identify safe parking zones; in dumping areas, they intelligently recognize unloading boundaries and surrounding safety risks; and on complex roads, they dynamically evaluate passing space, slope changes, and temporary obstacles.
Bronto has established a dual-wheel evolution system of "real data iteration + virtual scenario training." On one hand, the company continuously accumulates vast amounts of scenario data through real-world truck operations in mines, leveraging a closed-loop process of data governance, automatic annotation, model training, simulation verification, and real-vehicle testing to drive ongoing model optimization and upgrades. On the other hand, the company has developed an industry-specific "virtual driving school" for mines. Using virtual mining scenarios constructed by world models, it can replicate hazardous and complex conditions that are difficult to collect in reality, such as extreme weather, low nighttime illumination, dust obstruction, sudden road changes, and unauthorized personnel entry.
Chen Fangming stated that the core bottleneck in mine production efficiency lies in the level of coordinated scheduling across the entire transportation system. The value of Bronto's autonomous driving AI model lies in achieving a systematic leap from "single-vehicle autonomous driving" to "fleet-coordinated intelligent operations." Leveraging a supporting AI large-model scheduling system, it enables global capacity allocation, real-time optimal path calculation, dynamic task assignment, and real-time simulation of mine conditions, coordinating the entire fleet of unmanned mining trucks for efficient collaborative operations. This avoids inefficiencies such as empty vehicle travel, queuing, and process disconnections.
Bronto's "computing-electricity dual-drive" strategic framework and "mine AI agent" system aim to transcend the competitive dimension of single-device intelligence. The "mine AI agent" system comprises four core modules: the autonomous driving AI model serves as the decision-making hub, coordinating scenario understanding, path planning, global scheduling, and intelligent decision-making; unmanned electric mining trucks act as execution terminals, precisely completing transportation tasks and providing real-time field data feedback; the solar-storage microgrid ensures stable, low-cost clean energy supply; and massive real-world operational data continuously flows back for iteration, driving the entire system's ongoing evolution and upgrades. The "computing-electricity dual-drive" approach is the core operational methodology of the entire system. "Computing" encompasses the autonomous driving AI model, cloud-based training, vehicle-side reasoning, intelligent scheduling, and data evolution capabilities, forming the core of mine intelligent decision-making. "Electricity" covers photovoltaics, energy storage, electric mining trucks, and mine microgrids, forming the foundation for green mine operations. This system integrates mine energy flow, material flow, data flow, and decision flow into a unified framework, using electricity to drive equipment operation and support intelligent computing power, while leveraging computing power to optimize energy allocation and improve transportation efficiency.
Behind this strategic upgrade lies an evolution in Bronto's corporate value logic. The company has shifted from traditional hardware delivery of unmanned mining trucks to a full-chain output of mine AI system capabilities. It can provide customers with an integrated solution encompassing clean energy assurance, autonomous driving operations, intelligent scheduling optimization, data iteration upgrades, and long-term maintenance services, helping mining clients achieve multiple values such as cost reduction, efficiency improvement, and green transformation.
