search.noResults

search.searching

dataCollection.invalidEmail
note.createNoteMessage

search.noResults

search.searching

orderForm.title

orderForm.productCode
orderForm.description
orderForm.quantity
orderForm.itemPrice
orderForm.price
orderForm.totalPrice
orderForm.deliveryDetails.billingAddress
orderForm.deliveryDetails.deliveryAddress
orderForm.noItems
FEAT


ATURE


AUTOMOTIVE


ON THE ROAD TO FULL AUTONOMY Bielby A byRobert behindthetec


rt, seniordirector of echnolog


ect gy


utonomous cars are evolving from futuristic dreamto modern reality.


entirely, banishing drowsy, impaired and human motorists out of the equation Eventually, driverless cars will take


2017, nearly 40,000 people in the United distracted drivers fromroadways. In


States died on the roads. According to the National Highway Traffic Safety


percent of those accidents were due to Administration (NHTSA), about 90


For an automobile to be self-driving, human error.


perceive and act - sometimes in a split must be extensively trained to sense, the artificial intelligence (AI) network


relies onmyriad sensors on the vehicle, Understanding the driving environment second - in any traffic situation.


computer system. To accomplish this, the with data being processed by the car’s


autonomous vehicle will likely contain more lines of code than any other software platformcreated to date. By 2020, the typical vehicle will containmore than 300million lines of code, havemore than 1TB of storage, and requirememory bandwidth ofmore than 1TB per second. To make real-time decisions, a self- driving car’s AI systemrequires an


instructions. Self-driving vehicles exist uninterrupted streamof data and


driving the same route repeatedly. They today, but their success comes from


learn every detail of the route and


navigation system. This general reliance generate maps that informthe


geo-fencing - allows the vehicle to pay on recognising the route - known as


potential hazards. While geo-fencing attention to traffic, pedestrians and


may work over a limited route, an 18 MARCH 2019 | ELECTRONICS


recto ofautomotvesysy temsatMicronTe ogyandwhatwill it takebef


otivesyste foreyo canco


autonomous vehicle that relies on geo- fencing may not work well in unfamiliar places.


High-performance computers based on AI use deep neural network algorithms that allow autonomous cars to drive better than human-driven cars. A host of sensors work together to see the entire environment in 360 degrees, 24/7, at a greater distance and with a higher accuracy than humans can.


Imagine a car slamming on its brakes on a busy freeway. Through vehicle-to- vehicle and vehicle-to-infrastructure (collectively called V2X) communication, this event could be transmitted to oncoming cars, allowing themto avoid an accident.


SAFETY IN DETAIL The attention to redundancies de


signed into the hardware safety goes well beyond


systems to avoid bad decisions; it also includes infrastructure that allows self- driving cars to communicate with each other and their surrounding environment. This wireless interconnection, along with hardware redundancies, is governed by legislation that mandates safety requirements for differing levels of autonomy.


The NHSTA has established levels ranging fromLevel 0, meaning no automation, to Level 5, or full automation. The maj self-driving cars


Technology take eforeyouca commute towo gyasks,how ks tetoworkwithoutneedingtowatcwatchtheroa


owisadriverless ca ed


rive NEXT STEPS While the personal computer


historically drove memory technology, the automotive industry is likely to drive future memory technologies. For example, a vendor recently introduced an artificial intelligence platformfor autonomous driving, which is based on the industry’s leading-edge DRAM technologies. This platformdelivers more than 1TB per second of memory bandwidth to achieve Level 5 performance.


Micron has led the industry in both automotive memory solutions and graphics memories such as GGDR5 and GDDR6. The bandwidth associated with GDDR6 memories fa levels of autonomy i


n self-driving cars. cilitates higher


An autonomous compute platform that’s rich in memory bandwidth can evolve and refine self-driving algorithms. Over time, we will see improvements in algorithms. But those will be deployed as software upgrades, similar to how your smart phone receives regular updates.


are Level 2 capable and ajority of existing


based on computer hardware using relative mature and low-bandwidth memory devices. As driverless cars become more autonomous,memory technologies move fromthe back seat to the front to improve safety and performance.


The highest-performance memory technology available today, GDDR6 can operate in the high heat and harsh conditions associated with the automobile. In addition, GDDR6 is capable of fuelling AI compute engines that enable self-driving cars to address National Highway Safety Transport Association (NHSTA) and Society of Automotive Engineers (SAE) safety standards.


Micron Technology www.micron.com


www.micron.com/forms/contact-us / ELECTRONICS


car safer,what’s ro


fe oad?


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36