NXP Semiconductors is expanding its capabilities in the fast-growing autonomous vehicle space, giving Qualcomm executives a look at what the chip maker will get when it acquires NXP next year.
At the Electronica show in Germany this week, NXP official unveiled a new automotive radar microcontroller (MCU) that they said is four times faster than its predecessor in making safety-related decisions such as collision avoidance, lane changing and automatic braking. In addition, the company demonstrated a vehicle-to-everything (V2X) communications solution designed to enable the platooning of autonomous trucks.
The new MCU and truck platoon demonstration came days after NXP announced an extended partnership with Cohda Wireless in which NXP will use Cohda's V2X algorithms in its chipsets, which will making it easier for joint customers to accelerate the development of V2X and autonomous vehicle products.
The moves highlight NXP's strength in connected and autonomous vehicle technology, which the company bolstered last year when it bought Freescale Semiconductor for $12 billion. It's also one of the key drivers behind Qualcomm's bid to buy NXP for $47 billion, a deal that is expected to close next year.
The new MCU, the NXP S32R27, will increase the speed and accuracy of the safety decisions driverless cars will have to make on the road and improve the processing and analyzing of data coming in from the multiple on-board sensors designed to give the vehicles a 360-degree view of their surroundings and to track what's around them, from pedestrians to bicyclists. It's currently being sampled by top-tier automakers and will be generally available in the second half of next year.
The S32R27 is part of NXP's larger portfolio of products for connected and autonomous vehicles. The company offers a range of MCUs, single- and multi-core processors built on the Power Architecture that can be used as the basis for autonomous vehicle development. Earlier this year, NXP rolled out the Bluebox platform (pictured) for driverless cars that includes everything from the sensors that collect massive amounts of data to the compute engine capable of fusing all that data and ensuring that the right decisions are made.
In addition, NXP is partnering with DAF Trucks to develop the technology that will enable truck platoons to improve reaction times that will help lead to increased fuel efficiency, improve safety and reduce emissions. NXP, DAF and other companies before had used V2X technology and radar and camera systems to enable platoons of trucks to travel at almost 50 mph with 0.5 seconds between trucks, maintaining a distance of about 36 feet. The group, which includes the scientific organization TNO and vehicle company Ricardo, is now working on technologies that will reduce the minimum distance between trucks by 40 percent—to 0.3 seconds, or almost 23 feet at about 50 mph.
To reach this, the platooning systems within the trucks will have to react 30 times faster than a human driver, which will require communication between the trucks to happen in milliseconds, according to NXP officials. It will take such NXP technologies as MCUs, chips, power management integrated circuits (ICs) and networking components, most of which are found in the vendor's Bluebox platform. The MCUs will include the new S32R27, they said.
At Electronica, NXP and DAF are working with Honda and Siemens to demonstrate a platoon of autonomous trucks traveling through Munich, with communication between the trucks and intelligent traffic lights that adapt signaling based on the convoy's location and help improve traffic flow.
NXP officials last week said they are partnering with Cohda Wireless—which makes automotive communication software—to integrate Cohda's algorithms into the companies' joint RoadLink chipset.
"This software agreement will allow NXP to fully leverage the software-defined radio capabilities of our V2X platform," Torsten Lehman, senior vice president of car infotainment and driver assistance at NXP and a member of Cohda's board of directors, said in a statement. "Owning the firmware and adding NXP algorithms will further increase system performance as well as reduce development cycles for our joint customers."