With self-driving cars on the horizon, automotive design is one of the big topics at this year’s Design Automation Conference (DAC) in Austin. During Monday morning’s keynote, Lars Reger, CTO of the Automotive Business Unit at NXP Semiconductors, took the stage to talk about “The Revolution Ahead—What It Takes to Enable Securely Connected Self-Driving Cars.” Or, as he put it, “Why I think I have one of the coolest jobs in the semiconductor world at the moment.” NXP currently is the biggest automotive semiconductor supplier on the market, with more than 60 years of automotive experience and 2,400 automotive engineers. The disruption that self-driving vehicle technology is bringing to the industry is also presenting the semiconductor industry with tremendous opportunity. Reducing Human Error Via Autonomous Functions Today, 90% of the innovation in a vehicle comes via electronics, said Reger, noting, “We are converting horse carriages with combustion engines into robots that have wheels.” For Reger, the opportunity to use technology to create safer rides—significantly reducing the 90% of accidents caused by human error—would be the most rewarding outcome of all of these efforts. Autonomous functions, like automated acceleration, braking, and steering, as well as redundant sensor systems, such as radar, lidar, and sensor fusion, are paving the path to better vehicle safety and reliability. NXP Automotive’s portfolio for these self-driving robots include technologies that enable sensing, thinking, and acting. Reger calls this the “robustness tetrahedron,” explaining that it requires deep automotive experience in managing functional safety, functional security, device reliability, and road safety. The key enablers of securely connected self-driving cars are technologies that feed into the robustness tetrahedron: Radar sensing enables a 360-degree view around the car and also increases the amount of silicon in cars significantly. NXP has a postage stamp-sized radar system solution that combines NXP’s transceiver and Ethernet/CAN technologies with processor and power management solutions from Freescale, which NXP acquired last year. IEEE 802.11p enables wireless communications with high output power, supporting vehicle-to-vehicle (V2X) communications. Data security is another key criteria here. Processing all of the data collected by sensors calls for processors traditionally used in server farms, along with vehicle safety processors and sensor fusion technologies. Deterministic Ethernet provides the in-vehicle network, vital for self-driving capabilities with defined latencies and to support multimedia to sync audio with video. Security involving multiple layers of protection at different levels in the system can mitigate the risk that one component of defense will be compromised. Smart Trucks on the Road Many autonomous driving functions are already being tested. We’ve seen the news coverage on Google’s self-driving cars . In The Netherlands, the European Truck Platooning Challenge is bringing platooning of automated trucks to public roads. Having trucks equipped with smart technologies drive in columns, like in a caravan, can bring greater efficiencies, safety, and cost savings to transportation and logistics. Cars in 2020 may boast domain-based networks with separate branches for different areas, such as power train and driver assistance, all controlled by central gateways. For NXP, says Reger, the company is in a prime position to help accelerate the move toward self-driving cars with ready-made solutions providing the underlying technology. Christine Young![]()
![]()
↧