PORTLAND, Ore.--Parking assistance systems that give drivers a 360-degree view of obstacles around them are currently only available on high-end models, but Freescale Semiconductor Inc.--working with engineers at BMW--claims to have created a technology capable of reducing the cost enough to make them standard equipment.
"Freescale and BMW have cooperated on the definition of a new generation of Qorivva 32-bit microcontrollers, which we believe will enable surround-camera parking assistance systems to migrate to a broader range of vehicles," said Allan McAuslin, product marketing manager of safety and chassis microcontrollers at Freescale.
Key to the new microcontroller technology is its ability to compress the video from up to five cameras mounted on an automobile’s four corners--plus a backup camera--then stitch the images together in to a complete 360-degree image that can be displayed on the dashboard (usually in a donut-shape). Built on the Power Architecture, the new 32-bit Qorivva MPC5604E eliminates the need for costly individual video cables (at $10 each) by sending all image data over a single Fast Ethernet in-vehicle network.
Using compressed JPEG images, the Qorivva microcontroller is able to use a single cheap two-wire Ethernet cable to send the time-stamped video using the Autosar realtime operating system. As a result, BMW aims to debut the technology in its lower-end models. Freescale is also working with other automobile makers worldwide to assist them in switching to compressed video over Fast Ethernet, eventually hoping to make 360-degree parking assistance systems standard equipment on most models.
Freescale's new Qorivva 32-bit micro controller based on an e200 Power Architecture core has hardware JPEG video compression and Fast Ethernet to cut the cost of 360-degree view parking assistance.
This sounds like interesting technology. However, how safe is it for a driver to pay attention the road in real-time with the 360 view drawing away peripheral vision. Technology cannot ultimately replace the human driver's five senses that are important for driving. However, I'm wondering how well the implications of this technology would be for folks that need assistance. This may even be great for an aging population and education a younger population with safer driving habits. After all, technology this cool will hopefully make drivers more aware as it reduces "blind spots"--yet the video output may give a "video game-type" feel which could also be a hazard.
I wonder if it will pass for car insurance discounts because it is supposed to be an advanced safety feature.
There are many pros and cons and I would love to see more demos like the one you've provided. In addition, the image could some day be interactive and even be integrated into gps for smarter real-time navigation.
Similar to stitching operations performed by high-end applications like Photoshop for panoramas, the four images are aligned then distorted into a donut shape. Check out an example at: http://bit.ly/jAcyR9
David Patterson, known for his pioneering research that led to RAID, clusters and more, is part of a team at UC Berkeley that recently made its RISC-V processor architecture an open source hardware offering. We talk with Patterson and one of his colleagues behind the effort about the opportunities they see, what new kinds of designs they hope to enable and what it means for today’s commercial processor giants such as Intel, ARM and Imagination Technologies.