- Subaru aims to eliminate fatal accidents in its vehicles by 2030 through AI integration.
- The company established an AI lab in 2020 to accelerate safety tool development.
- Generative AI is being integrated with existing machine learning tech for EyeSight enhancement.
- EyeSight, introduced in 2008, includes features like adaptive cruise control and lane departure warnings.
- Subaru collaborates with Google Cloud for deep learning research and AI model development.
- A hybrid model is utilized, combining Google Cloud for image analysis with Asura Net for tasks like object identification.
- Robust on-premises infrastructure, including Dell PowerScale NAS systems, manages vast datasets.
- Partnership with AMD promises enhanced AI inferencing capabilities with the Versal AI Edge Series Gen 2 SoC.
Main AI News:
Several years ago, Subaru embarked on an ambitious mission: to eradicate fatal accidents involving its vehicles by the year 2030. To achieve this lofty goal, the Japanese automaker is placing a significant emphasis on harnessing the power of artificial intelligence (AI).
In 2020, Subaru established an AI lab dedicated to accelerating the development of safety tools aimed at enhancing the overall safety of its vehicles. As the AI market continues to evolve and mature, Subaru has been progressively expanding its arsenal of cutting-edge technologies.
During a recent interview at the Dell Technologies World conference in Las Vegas, Takashi Kanai, deputy chief of Subaru Lab and manager of the automaker’s advanced driver assistance systems development department, shed light on Subaru’s AI-driven initiatives. Kanai highlighted the integration of generative AI alongside the existing machine learning technologies employed by the company.
At the core of Subaru’s innovation lies the relentless pursuit of enhancing its EyeSight driver assistance system. First introduced in 2008, EyeSight is a comprehensive safety feature integrated into select Subaru models. It encompasses a wide range of functionalities, including adaptive cruise control, lane departure warnings, automatic braking, and blind-spot detection. By incorporating generative AI, Subaru aims to further augment the capabilities of EyeSight, pushing the boundaries of automotive safety technology.
Since its inception, EyeSight has garnered widespread acclaim, with over 5.5 million Subaru vehicles equipped with this advanced safety feature. Notably, Subaru’s collaboration with Google Cloud has been instrumental in leveraging managed services for deep learning research and AI model development.
Subaru adopts a hybrid model, combining the capabilities of Google Cloud for image analysis with its proprietary Asura Net neural network for tasks such as object identification and semantic segmentation. Moreover, Subaru’s robust on-premises infrastructure, bolstered by Dell PowerScale NAS systems, enables the efficient management and utilization of vast datasets generated during the testing and development phases.
Looking towards the future, Subaru’s strategic partnership with AMD heralds a new era of optimization in AI inferencing capabilities. The upcoming Versal AI Edge Series Gen 2 SoC, slated for integration into future iterations of EyeSight, promises to deliver enhanced recognition processing, reduced latency, and cost efficiencies.
As the automotive industry embraces the transformative potential of AI-driven technologies, Subaru stands at the forefront, committed to redefining the standards of safety and innovation. With each advancement, Subaru reaffirms its dedication to realizing a future where fatal accidents are relegated to the annals of history.
Conclusion:
Subaru’s strategic investment in AI-driven safety technologies, as evidenced by its integration of generative AI into the EyeSight system and partnerships with industry leaders like Google Cloud and AMD, underscores its commitment to innovation and safety. This not only positions Subaru as a frontrunner in automotive safety but also signals a broader trend within the market towards AI-powered solutions for enhanced driver assistance and accident prevention. Competitors may need to accelerate their own AI initiatives to remain competitive in this rapidly evolving landscape.
