The government-funded research institution ETRI (Electronics and Telecommunications Institute) has announced a breakthrough in its development of image recognition processors geared towards driverless car technology. (Image: ETRI)
DAEJEON, Sept. 20 (Korea Bizwire) – The government-funded research institution ETRI (Electronics and Telecommunications Institute) has announced a breakthrough in its development of image recognition processors geared towards driverless car technology.
The achievement – the successful design of a nine-core processor – is a notable one for South Korean interests, as it was only last year that the quad core processor labeled Aldebaran was developed in-house. Prior to that, processor manufacturing had been reliant on foreign designs.
Measured at 7.8 mm wide and 6.7 mm long, the processor is comparative in size to a human fingernail. Despite its paltry physical parameters, successful testing has proved the processor’s improved image recognition capabilities, real-time UHD footage support and radar and GPS signal detection.
The engineering team expects its product to be used in ultrasounds and Lidar (Light Detection and Ranging) technology.
Besides its core functions, the processor can also store video footage like a black box, and can run autonomous vehicular system checks that will reportedly identify malfunctioning electrical equipment with 99 percent accuracy.
The engineering team expects its product to be used in ultrasounds and Lidar (Light Detection and Ranging) technology. (Image: ETRI)
The ETRI will share its processor blueprints with a manufacturer by the end of the year, and is aiming to release the product into market by next year.
Regarding the successful development of the nine-core processor, ETRI researcher Kwon Yong Soo said, “The pre-processing of sensors that go into standard autonomous cars required additional cores, the efficiency of which the resulting ‘one-chip’ technology has significantly improved.”
Speaking on the future goals of the ETRI, Kwon said, “In the long term, the goal is to develop a chip that can autonomously drive towards a destination that has been decided through a conversation between human and machine.”
S.B.W. (sbw266@koreabizwire.com)
