Mobileye EyeQ Software Development Kit provides new flexibility for future software-defined vehicles
Mobileye has launched the EyeQ Kit, the company’s first software development kit (SDK) for the EyeQ system-on-chip (SoC) that powers driver-assistance and future autonomous technologies for automotive manufacturers worldwide. Designed to make use of the powerful and highly power-efficient architecture of the EyeQ 6 High and EyeQ Ultra processors, the EyeQ SDK allows manufacturers to use the EyeQ core technology whilst deploying their own differentiated code and human-machine interface tools on the SoC platform.
According to Prof. Amnon Shashua, Mobileye president and chief executive officer, the EyeQ Kit allows users to benefit from the best of both worlds, the core technology and their own expertise in delivering unique driver experiences and interfaces.
“As more core functions of vehicles are defined in software, we know our customers will want the flexibility and capacity they need to differentiate and define their brands through code,” he says.
Customising vehicle functions
Through the EyeQ SoC hardware and its software, car makers have access to a broad set of Mobileye technology, including computer vision, “REM” crowdsourced mapping and RSS-based driving policy. By further using the EyeQ Software Development Kit, these manufacturers can also make further use of the power of Mobileye’s system-on-chip to augment the advanced driver functions with a look and feel that is unique to their particular brands. And as the visual demands for interaction and communication between drivers and vehicles grows more complex, the EyeQ SDK gives manufacturers a new path to tailor critical information flows within the vehicle. The EyeQ SDK can be used to help support features such as surround visualisation, automated lane-keeping and road-sign recognition through more advanced augmented reality displays.
The EyeQ Kit was developed by the Mobileye team of engineers, each drawing on the combined long experience it has across Intel and Mobileye products, with experts in compilers and development environments such as OpenCL standards, making use of compilation frameworks used for intensive computing and deep learning. This approach allows the end users of the system to develop their own applications easily and efficiently. It also opens the ability to co-host third-party applications, which lowers the costs of integrating other chips.
An Approach to Product Differentiation
From general-purpose CPU cores to high density computing accelerators – including deep-learning neural networks – EyeQ has a scalable and modular architecture that seeks to achieve high performance while offering a suitable power efficiency to deploy artificial intelligence at the edge for automotive applications. Previous generations of the EyeQ chip have been deployed in more than 100 million vehicles, with dozens of the world’s largest automotive manufacturers using them to provide safety and driver-assistance features in hundreds of models worldwide.
Now, the EyeQ SDK aims to reduce development costs, accelerate time to market and enable hardware vendor flexibility for the full development cycle – from conception to deployment and performance tuning, providing the product differentiation opportunities that the manufacturers want to achieve.
According to Shashua, Mobileye’s core technologies of computer vision, driving policy, REM and RSS are driven by purpose built SoCs allowing for scale and efficiency. Customers can now build on top of Mobileye’s core technologies while benefiting from its diverse set of accelerators purpose built for ADAS and AV.
“Automotive manufacturers can rely on the EyeQ SDK and EyeQ processor family to bring a technologically advanced vision for their brands to life, quickly and efficiently,” he concludes.
- UK manufacturing steps up to COVID-19 crisis - April 2, 2020
- Clustering Innovation - March 12, 2020
- A Global Monitor - March 6, 2020