Page 806 - AI for Good Innovate for Impact

AI for Good Innovate for Impact

language input via voice. This significantly reduces task execution time from minutes to seconds,
streamlining the overall driving experience.

Problem to solve: Conventional vehicle control configuration is time-consuming and unintuitive,
often requiring multiple manual inputs across isolated systems. These processes typically take
several minutes and demand technical knowledge, posing usability challenges for non-expert
users.

Limitations of existing solutions: Existing solutions rely heavily on static presets or manual
configuration interfaces, offering limited adaptability to user preferences or dynamic
environments. They lack the capability to orchestrate functions across subsystems in real-time,
and often cannot integrate seamlessly with legacy vehicle platforms or third-party services.

Benefits and drawbacks of the AI-based approach: The system leverages a large language
model trained on approximately 20 million automotive tokens and 400,000 task-specific
instructions to parse user input and autonomously generate scenario logic chains comprising
preconditions (e.g., vehicle speed < 30 km/h), triggers (e.g., voice command detected),
and executions (e.g., activate low-speed cruise mode). This enables context-aware, modular
orchestration of complex vehicular functions.

Benefits include:
1. High-performance natural language understanding for domain-specific queries.
2. Dynamic invocation of cross-system functions (e.g., ADAS, infotainment, HVAC).
3. Integration with SparkLink to enable low-latency, high-reliability short-range
communication.
Limitations include:

1. Dependence on training data quality poses risks for edge cases and uncommon driving
scenarios.
2. Ongoing challenges in model interpretability, error recovery, and adaptation to real-time
edge environments.

Impact on Intelligent Transport
This system supports global efforts toward sustainable development by promoting more
efficient, eco-friendly, and innovative transportation solutions. By enabling flexible scenario
orchestration and automated vehicle control, it significantly reduces energy waste, enhances
vehicle usage efficiency, and promotes environmentally responsible travel behavior.

The integration of advanced artificial intelligence models with digital twin technology allows
for real-time optimization of resources and decision-making. This leads to a safer, more
comfortable, and sustainable driving experience. Modeling entire vehicle services in the cloud
for training and testing AI systems strengthens the foundation of smart transportation networks.

These innovations contribute to building sustainable and resilient urban mobility, supporting
climate-conscious strategies, and fostering technological advancement. Ultimately, this
approach enhances the quality of life in cities and promotes the broader vision of intelligent,
green, and inclusive transportation systems.

Partner name: Ali Cloud Computing Co. Ltd. (Collaborating on LLMs technologies), IEEE SA
(Collaborating on the standard for MLLMs platforms)

770