ZF chassis boss sees on-board processor "bottleneck"

RSS_Auto_Poster

Well-known member
Written by Ian Adcock
7479.jpg

ZF"s takeover of TRW in 2015 sent shockwaves through the industry by over night creating a mega global business with the industrial and financial clout and electronics capability to challenge the top Tier 1s. Some observers might have questioned the takeover"s logic or whether the cultural differences might prove destructive, as they were in the Daimler-Chrysler "merger of equals." But Dr. Holger Klein, head of ZF"s Car Chassis Technology group, sees nothing but benefits from two businesses sharing similar visions joining forces. "At the end of the day the rationale behind the deal was industry logic where you say, I want to offer the complete chassis" and missing components on our side were complemented by components and know-how from TRW," he said during a recent interview with Automotive Engineering. "The vision is the same, centered around safety, efficiency and autonomous driving. The components were very complementary." ZF could have bought smaller companies with similar expertise, but the TRW opportunity was "a once in a lifetime opportunity" for a mighty technology synergy including in Dr. Klein"s chassis systems domain, a 6B business with 16,000 employees. He said all of the division"s products including suspension, linkages, structural parts, ball joints and stabilizers are becoming "intelligent" mechanical components. "What we say is that the chassis controls the vehicle"s lateral, longitudinal and vertical dynamics as well as its guidance. If you want to be successful in this area and be a technology leader you really need to combine these components into mechanical systems," he noted. The processor-speed challenge ZF"s acquisition of TRW expanded its technological capability "enormously," said Klein, "If you "peel the onion" and talk about the chassis technology itself, we lost the steering systems joint venture with Bosch but we got steering technology from TRW which, by the way, is very much integrated into the business. The braking side is very important for longitudinal dynamics and then this entire arena of ADAS [advanced driver assistance systems] and active safety which helps us a lot with sensor technology like cameras, radar, etc. This all comes in together in what we need to serve the mega trend of autonomous-driving cars." It is this melding of electronics with mechanical and hydraulic systems that is, arguably, one of the big challenges facing the industry. Dr. Klein, however, argues that it"s the computing power on-board the vehicle which is proving more of a bottleneck than the other systems. "We have to develop strategies to counter this," he said. Computers need to translate image recognition and calculate trajectories to define a strategy that can prime the actuators ahead of any eventuality. "We"re talking milliseconds here and speeding up actuators through electrification, like a braking e-booster or yaw control, will improve reactions even more," Dr. Klein explained. Clearly the emergence of Artificial Intelligence will be key to future autonomous driving strategies as ZF"s deal with Nvidia to develop and produce the ZF ProAI processor from 2018 onwards demonstrates. "There always was a discussion of how many computers or ECUs we need in a car," he said, "and what do you integrate into the ECUs and what kind of computing power do you need? What we see is that we"re already reaching certain limits of our computing power if the car is in an emergency situation. Do you need to ramp down the multi-media system just to get a few more milliseconds of computing power to solve these complex algorithms? There"s a clear need for more powerful platforms than we have today." Preparing for X-by-wire An intrinsic element of this development will be the roll-out of 48-volt hybrid systems as an enabling technology for both autonomous driving systems and other technologies within vehicles that are currently mechanically or hydraulically powered. "We actively see the electrics driving us to the next generation of actuation which will certainly be X-by-wire," Dr. Klein forecast. "The challenge will be how do you create redundancies and show the fail safe mechanisms and make sure the electrics are really proven? There"s also a need for legislation to be updated to allow for these developments to happen." Whilst he concedes that electromobility "helps" with city driving in lowering CO2 and other emissions, Dr. Klein believes the easiest solution is a hybrid. "I think there is no way round hybrids, although we are all engineers and we doubt if that is the most efficient means of solving the issue." He concedes that having two propulsion systems on board is not a thermodynamic optimum, nor is it a mass and packaging ideal.   "The big challenge is which technological solutions will deliver the right range, meet customer demands and is CO2 neutral?" he said. Underlying this growing sophistication and complexity is "enormous" cost pressure, which Dr. Klein believes global legislators don"t always consider. Shifting from one technology to another overnight presents "a huge challenge" to an industry that makes decisions that won"t materialize for five years and won"t be amortized for another 15 years, he said.



Date written: 17-Apr-2017 12:48 EDT

More of this article on the SAE International Website

ID: 7479
 
Back
Top