No, you won't hear that at an Indianapolis 500 race event any time soon. But there's little doubt that electronics technology stands to permeate even further throughout all kinds of automobiles of the future, from compacts to race cars, translating into millions if not billions of semiconductor ICs.
Scads of sensors, processors, DSPs, memories, ASICs, transceivers, LEDs, flat-panel displays, power ICs, solid-state relays, laser diodes, audio ICs, cameras, data-conversion devices, and radar gear are sketched into drawing-board designs. As a result, cars will be safer, more convenient to operate, easier to maintain, and a pleasure to drive. Semiconductor IC manufacturers also are delighted as they prepare for the huge market predicted by most analysts.
For example, Allied Business Intelligence forecasts that the worldwide automotive semiconductor market will grow from 2002's $12.4 billion to $17 billion by 2007. Strategic Analysts, another market research firm, predicts that electronic systems in cars will account for more than 30% of a typical car's cost by 2008, up from today's 20%.
A taste of things to come can be seen at the many automotive expositions where car makers showed off not only their newest models, but concept vehicles as well, many chock full of technology innovations (Fig. 1). Just a short list of coming features shows multi-speed electronically controlled transmissions with finger-operated tap starting, adaptive electronically controlled suspension systems, 3D topographical mapping and navigation systems, and infrared cameras for under-chassis displays. Electric motors are definitely coming in hybrid vehicles to make driving less expensive and more environmentally friendly. So are handheld "key fob" information technology systems that allow drivers to "personalize" their vehicle's various settings to individual preferences. Then, there are computer-assisted parking, adaptive cruise control, and exterior-perimeter lighting.
Telematics capabilities will expand to include automated vehicle service notifications and appointments, VHF radios, flip-down ultra-thin flat-panel displays for receiving and viewing satellite movies and videos, satellite phones, video games, computers, Internet access ports, and digital video recorders with satellite-transmission capability. On the safety and anti-theft side, smarter air bags, radar accident-avoidance systems, and biometric ignition and door locks will become standard features.
Many of these technological features will be government-mandated, driven by the need to make cars more fuel-efficient, less polluting, and safer to drive, as well as to diminish driver distraction. The U.S. National Highway Transportation and Safety Administration (NHSTA) estimates that 20% to 30% of all car crashes are caused by driver distraction. Another NHTSA document, "Vehicle Safety Rulemaking Priorities, 2002-2005," indicates that future legislation will likely address the issue of vehicle roller and driver distraction.
Such government mandates translate into more opportunities for electronics technology to be embedded in cars, with sensors of all types—particularly microelectromechanical-system (MEMS) sensors—making huge inroads.
One primary safety issue is tire-pressure monitoring, which was to be mandated for new passenger cars beginning late last year. Under the Transporta-tion Recall Enhancement, Accountability and Doc-umentation (TREAD) Act issued by the NHTSA, U.S. automakers must install tire-pressure monitoring systems on at least 10% of their 2004 model passenger vehicles and light trucks. This is supposed to rise to 35% in 2005 and 65% in 2006, and thereafter in all vehicles. Auto industry analyst J.D. Powers & Associates predicts that more than 17 million tire-pressure monitoring systems will be in cars by 2007.
The NHTSA is putting the finishing touches on regulations that would add a tire-pressure monitoring system to every tire. The agency's TREAD Act mentioned earlier makes provisions for two kinds of tire-pressure monitoring methods—indirect and direct. In the lower-cost indirect approach, the system piggybacks atop an existing vehicle's anti-lock braking system (ABS). Special ABS software algorithms determine if one wheel is rotating faster than the others, a sign that the wheel's diameter is smaller and thus under-inflated. The indirect method doesn't know anything unless the wheels are rotating, nor does it know anything if the tires lose pressure at similar rates.
A direct system, which incorporates a 4-bit microprocessor, an RF transceiver, a sensor, and about 4 kbytes of ROM in every tire, can accomplish all of this and provide additional information like tire temperature. But it also costs about five times as much. While chip suppliers and some tire manufacturers favor a direct system, auto manufacturers have yet to back it. This is because of its higher cost and the concern that chip makers may not have the capacity to ramp up production levels to quickly supply the millions of devices that would be needed.
For now, chip suppliers are working with system-level OEMs and tire manufacturers to show that the direct approach is feasible (Fig. 2). It looks as though a middle-of-the-road approach may be taken by auto manufacturers to phase in the direct approach over time, putting the more expensive but higher-performance tire-pressure monitoring systems in high-end cars initially.
STANDARDS NEEDED With the deluge of electronics in automobiles, standards must be instituted so that electronic systems can communicate with one another and over specific architectures. Even the issue of using 42-V systems instead of the present 12-V systems has yet to be settled, but there's a clear move toward a 42-V standard.
To promote better standardization, the Flex Ray Consortium of automobile, IC, and automotive-system manufacturers (formed in September 2002) is pushing to develop an open standard for high-speed automobile bus systems, such as the X-by-wire architecture. Last year, German automakers and electronic suppliers formed a consortium that aims to put a standard software vehicle infrastructure on the road by 2006. BMW, Bosch Automotive, Continental Automotive Systems, DaimlerChrysler, Siemens VDO, and Volkswagen intend to create a software foundation that serves every electronically controlled component in an automobile.
Known as Austar, for "automotive open systems architecture," the initiative could lead to a plug-and-play architecture that will help cut down the scores of microcontrollers in future cars, from the present 40 to 60 to about 20 (Fig. 3). Although largely German, the Austar consortium aims to make it international. In fact, General Motors has indicated interest in joining the consortium.
But not everyone is convinced that a standardized software architecture will be workable. Some validly point out that neither embedded software nor firmware are pure software, and they may not be amenable to a plug-and-play environment. Still, the increasing penetration of electronics into automobiles is practically synonymous with software, and many software experts warn that software complexity in cars, in terms of lines of code written, will rise 100-fold. This situation requires some sort of software standardization if electronics technology is to successfully make it into the automobile of the future.
One trend is clear and unstoppable. Future automobiles will gravitate toward "fly-by-wire" control systems, using electric motors to control braking, steering, and suspension functions, instead of hydraulic actuators and electromechanical linkages. Evidence of this can be seen from the new types of advanced and computationally intensive microprocessors that chip makers are preparing to form the basis of a computing foundation for future cars.