"We think there is a strong value proposition around putting intelligence in electronics as opposed to hardware," says Mike Crane, Siemens VDO North American Director of Powertrain Gasoline Systems (www.usa.siemensvdo.com). The importance of electronics under the hood is something that is being driven by a number of factors, both market-driven and regulatory. Whether it is the seemingly ever-increasing price of fuel, the attention being paid to alternatives such as ethanol, hybrid technologies, or environmental regulations, what Crane says is happening is a shift in fundamental engine architectures in order to accommodate them. Joe Fadool, Siemens vice president, Powertrain Electronics and Drivetrain Systems, adds, "It's less expensive to control functions better, and more tightly, using electronics and software rather than adding after-treatment systems or mechanical control systems." In other words, it is about controlling things via electronic advances.

Consider, for example, the potential proliferation of diesel engines for passenger cars. Fadool sees the need for electronic control systems to help manage the emissions from those engines. Consider the growth of boosting--superchargers or turbochargers. "It's no longer a turbocharger with a bypass valve or a waste gate," says Crane, "it's a variable-nozzle or a variable-geometry turbocharger. You have to map the turbocharger and control that map." Valve disablement systems in order to shut off cylinders to minimize fuel use? Again, something that needs to be managed electronically.

Hybrids, of course, put plenty of electronics under the hood. "Clearly, hybrids require not only additional control," says Fadool, "but power electronics--high temperature, high-power electronics--to control the electric motors." Suddenly there are inverters involved in the motor-generator setup. And there are potential demands to control 80-kW motors for hybrid vehicles. While he admits that it is still pretty much a niche market today, he says that they're estimating as many as a million units being sold by 2012. (He notes, "One advantage that Siemens has is that we do inverters and high-power motors for industrial applications." This know-how is being transitioned to the automotive side of the business.)

One thing that Fadool says they're working on is to actually use software in place of physical sensors on engines. He explains that they're developing mathematical functions that can be used to infer what is going on and to respond accordingly. That is, they're taking information that is being provided by sensors on the engine, then figuring out whether based on parameters (e.g., air flow, throttle position, spark advance) it is possible to determine what the appropriate response would be. Another example is instead of having a physical sensor on electromagnetic valve actuators, signals from the electromagnetic field generated could be used by software to "sense" the position of the valve without a physical sensor.

This is not to say that there will necessarily be the elimination of sensors and actuators. In fact, Crane says that there will undoubtedly be the use of "smart actuators." Rather than having the ECU deal with functions, the electronics will be right on the actuator so that control can be local. Consequently, this opens up space in the ECU to deal with complex functions. Fadool points out that while it is true that there is a lot of computing power in a given vehicle that spends comparatively long periods of time idle or underutilized, "Under the hood, that's not necessarily the case because the engine is a real-time, embedded system." Things are being controlled at the micro- or millisecond level. So the 32-bit processors in the ECUs are hard at it.