Electronics Weekly Electronics Design & Components Tech News
“Fixstars possesses both software for deep learning and optimisation technology that allows more efficient utilisation of hardware,” said Renesas automotive v-p Takeshi Kataoka. “I am confident that our collaboration will enable us to provide support for software development optimised for automotive applications.”
The Automotive SW Platform Lab will start operation in April, developing ways for deep learning and operating environments to continually update learned network models to manage recognition accuracy and performance.
“After developing a deep learning application, it is not possible to maintain high recognition accuracy and performance without constantly updating it with the latest learning data,” said Fixstars CEO Satoshi Miki. “Fixstars plans to focus on these machine learning operations for the automotive field, as we work together with Renesas to develop a deep learning development platform optimised for Renesas devices.”
As part of the collaboration, the partners have launched ‘Genesis for R-Car’, a tool that aids the selection of Renesas’ processor ICs for driver-assistance and autonomous driving applications, using Fixstars’ Genesis cloud-based environment.
The tool models processing execution time in frame/s and the recognition accuracy of Renesas’ V3H convolutional neural network (CNN) accelerators on sample images using generic CNN models such as ResNet or MobileNet. A service allowing customers to use their own CNN models for evaluations is planned.
“It also allows engineers to select the device and network they wish to evaluate and perform operations remotely on an actual board,” said Renesas. “Engineers can use the environment to confirm evaluation results in tasks such as image classification and object detection, with the option to use their own images or video data.”
