Unmanned Systems Technology 002 | Scion SA-400 | Commercial UAV Show report | Vision sensors | Danielson Trident I Security and safety systems | MIRA MACE | Additive manufacturing | Marine UUVs
12 Platform one Spring 2015 | Unmanned Systems Technology The increased demand for processing power in automotive ADAS (advanced driver assistance systems) applications for driverless cars is having a profound effect on Freescale Semiconductor’s multicore roadmap. The i.MX6 family has seen great success in a wide range of applications, ranging from single-core to quad-core implementations, but this is changing, according to Geoff Lees, senior vice- president and general manager of the microcontroller division at Freescale. Instead, the i.MX7 range will focus on general consumer applications with single- and dual-core devices using the ARMv7 architecture, while the i.MX8 will focus on the automotive sector with larger quad and octal devices with 64-bit ARMv8 architecture such as A53 and A57. These will be coupled with more power- efficient Coretex-M controller cores. All of this will need a new programming approach, Lees said, and it is committing to the OpenCL language to do so. OpenCL is a higher level programming language that is open rather than tied to a particular architecture, and is aimed at multicore systems with elements for detecting which resources are available for running code. “Our i.MX roadmap is all about OpenCL,” said Lees. “The i.MX 7 has morphed into low-power, low-cost consumer, non-automotive applications with a power consumption of 800 mW and a target power of 500 mW and lower. The i.MX8 is the focus for the quad and octal devices for automotive infotainment and cockpit imaging with a pure v8 A core with ARM-Cortex M sub- processing.” This will also see a massive step up in GPU processing with the Vivante GC700 cores across the range. “We will implement massive arrays in the i.MX8 family – automotive is a key driver and there’s a lot of graphics and image capture applications to follow,” said Lees. Autonomous vehicle driver Software The vehicle-to-infrastructure technology is a key element for driverless cars and is powered partly by an i.MX6 processor from Freescale Semiconductor
Made with FlippingBook
RkJQdWJsaXNoZXIy MjI2Mzk4