big-ideas-features
 

Java Under the Hood

Perrone Robotics puts Oracle technology on the dashboard of Neil Young’s zero-emissions vehicle.

by Margaret Harrist, November 2010

The freedom of the open road is an integral part of the American Dream. Although the spike in gas prices in 2008 got many people interested in smaller, fuel-efficient cars, large trucks and roomy sedans continue to be the top-selling vehicles in America. But what if it were possible to heed the call of the open road in a vehicle that was even bigger and heavier than many sold today—but used far less energy and produced little or no emissions? Several years ago, Rock and Roll Hall of Famer, inventor, and car enthusiast Neil Young decided it was time to get serious about exploring how to achieve that goal.

Young assembled an enterprising team of engineers, mechanics, and technicians from across the country to build LincVolt, a sleek 1959 Lincoln Continental convertible transformed into a long-range electric car supported by an onboard generator. The goal: to create clean automobile propulsion technology that serves the needs and spirit of today’s driver but reduces the demand for petroleum and has zero emissions—and possibly eliminates roadside refueling entirely.

Although this experimental vehicle has undergone several engine changes since the project began, Java technology has been key to the team’s progress and is a central component of LincVolt’s mission-critical control and monitoring systems. 

Java on Board

One of the key technologists Young tapped for the LincVolt project was Paul Perrone, founder of Perrone Robotics. Perrone had helped develop an autonomous robotic car that had competed in the Defense Advanced Research Projects Agency (DARPA) Urban Challenge in 2007 and the DARPA Grand Challenge in 2005. Perrone’s knowledge and experience with building and integrating telemetry made him a perfect addition to the LincVolt team.

“Initially they wanted to add some sensors to the car to gather statistics about its energy efficiency,” Perrone says. “I knew we could set that up quickly using our MAX platform, a suite of software services based on Java technology that is designed specifically for robotics and automation.”

Perrone started on the sensor telemetry project in the summer of 2008 and had the system up and running in a short time. With that success, his involvement in LincVolt grew.

“The car has all these disparate components—an electric motor that gets power from the batteries; the turbine that powers the high-voltage DC bus, which charges the 104 3.2-volt batteries; and other onboard components that had never been hooked together before. They needed a way to control these things and orchestrate how they operate, so our Java-based robotic controls were the answer,” Perrone explains.

When Young turns the key today, the Java-powered control computer comes alive and begins turning on the engine components. Once the car is in motion, the turbine boots up and starts charging the batteries. The LincVolt Intelligent Dashboard provides the driver with information about the car’s speed, kilowatts, controller’s temperature, bus voltage level, battery level, and more.

If there’s an issue, a red light starts to flash. If it’s a serious issue, the red light goes solid. Young can then take a look at the seven-inch touchscreen in the dash to scroll through the warnings and see exactly what’s happening.

“For example, our monitoring and sensing system communicates if any of the batteries under the hood are overheated or over voltage—and tells us exactly which battery is the problem so we can perform pinpoint diagnostics,” Perrone says.

With the exception of the touchscreen, the dashboard of the LincVolt is true to its 1959 roots. The entire car is a pristine classic. When it passes you by, the only hint that this is no ordinary car is the silence of the electric engine. 

Under the Hood

LincVolt has gone through a number of iterations since Perrone joined the team, which also includes Johnathan Goodwin, an expert in diesel engine conversion and owner of H-Line Conversions; UQM Technologies, which is supplying the electric motor; and Capstone Turbine, which is building the car’s onboard generation technology.

“LincVolt is really a rolling lab,” Perrone says. “The team experimented with generating hydrogen from water onboard, but that’s been put aside in favor of Capstone’s 30-kilowatt, biodiesel-powered turbine.”

A renewable alternative fuel produced from a wide range of vegetable oils and animal fats, biodiesel was shown to significantly decrease emissions of particulate matter, carbon monoxide, and hydrocarbons in a 2002 study by the Environmental Protection Agency. Later studies indicated that biodiesel particulate matter is also less toxic than diesel.

The new Capstone turbine should maintain a charge on the batteries for the life of a good road trip, enabling the kind of nice, long ride that the Lincoln Continental was designed to provide—but without the gas station stops, heavy emissions, or engine noise.

Additionally, the intention is for the turbine to have enough capacity not only to get through a long drive but to still have the power afterward to generate energy for the home when it is parked. And because the average home consumes about 30 kilowatt hours per day, perhaps the LincVolt engine could enable people to generate all the energy their homes need. 

Taking Controls to the MAX

According to Perrone, the development of the onboard generator resulted in some pretty novel innovations, and the LincVolt team is pursuing patents for that technology. However, the parts that make up the car are off the shelf, making LincVolt a mechanical integration project.

Throughout the building and testing processes, Perrone’s Java-based MAX technology platform has enabled sensor telemetry, controlled mission-critical systems, monitored performance of each component, and pushed all this information to the Web so the team can observe the data and the growing number of online LincVolt fans and followers can see the progress. Additionally, an onboard camera enables virtual friends of the project to ride along during test-drives.

“The MAX technology component of LincVolt has been developed very modularly,” Perrone says. “When the team switched out generators in the various phases, a lot of our software didn’t need to change. We had to do a little configuration and some minor customizations, but 80 to 90 percent of it has stayed the same because the modules are flexible. In keeping with the spirit of the project, they are also general enough that they can apply to other projects as well.”

MAX started out as a research project nine years ago. It initially included general-purpose software libraries for sensors and actuators that Perrone implemented for some hobby robots. He and his team then started adding configurable software libraries for more-complex operations and eventually developed a number of profiles for the patent-pending platform.

Today, Perrone Robotics licenses the MAX platform and toolkit for robotics and automation applications.

“Java provides the maximum flexibility across underlying hardware, operating systems, and integrated third-party components, leaving the developer free to focus on the unique creativity essential for truly innovative projects like the MAX platform,” says Hasan Rizvi, senior vice president of product development at Oracle. “It provides a portable platform that makes it faster, easier, and cheaper to develop cutting-edge applications.”

Those capabilities have been key for Perrone’s company. “In robotics, it’s like the Wild West. There are very few real standards that are used, so there are a variety of platforms, systems, and microcontroller environments. Java has enabled us to port across platforms easily and scale from the smallest hobby robots to large-scale implementations,” Perrone explains.

Additionally, the simplicity of the Java programming language has been an advantage. “It’s very intuitive, and we can leverage a wide range of third-party commercial and open-source software libraries and tools. That helps when we’re bringing people in to work on a project. For us, it’s been the perfect programming storm,” he adds. 

Putting It On the Street

In 2005 Perrone Robotics put its technology to the test in the DARPA Grand Challenge, a race of autonomous ground vehicles across the Mohave Desert. Perrone entered the race to demonstrate a faster and cheaper approach to complex robotics development using the MAX platform. Up against million-dollar projects that involved 100 man-years of work, the team spent US$60,000 and one or two man-years—and was one of 40 semifinalists invited to trials at the California Speedway.

The company then began to develop vertical frameworks and software drivers for different types of sensors—some of which were specifically designed for unmanned ground vehicles. The company also built an unmanned air vehicle, and it has leveraged MAX technology for several factory automation projects involving laser technology.

The experience Perrone’s team gained through these projects led to one of Perrone Robotics’ recent large-scale projects with the Pennsylvania Turnpike. At JavaOne 2010, Perrone presented a case study explaining how his company used a combination of the Java-based MAX Vehicle Measurement Framework and MAX robotics/automation platform, Java Real-Time System, the Oracle Solaris operating system, and Sun Fire servers with Intel Xeon processors to create a laser-based, real-time vehicle measurement system. This system classifies vehicle size, calculates speed, notes lane position, takes photos of license plates, and weighs each vehicle through scales in the road—all in real time and while the vehicle is traveling at normal speeds.

“This laser system, which was developed by SICK (a global manufacturer of sensors for factory and logistics automation), is far more efficient than current traffic monitoring technology, which tends to be costly, difficult to maintain, and provides very limited data,” says Perrone.

“We have a long history of success using Java for deterministic, real-time projects, embedded systems, and microcontrollers,” he says. “These aren’t just experiments, but systems that are working and running and have true mission-critical roles—and we’re able to do it at 1/100th the cost of competing solutions in some cases.” 

Back in the Garage

Once the new biodiesel-based turbine is installed in LincVolt, Perrone will fine-tune the configurations and work with the team to integrate it into the engine.

Every tweak puts the team closer to proving that the technology exists to produce an energy-efficient, zero-emissions vehicle that’s big enough for the American lifestyle.

Looking ahead, the LincVolt Intelligent Dashboard might assume some autonomous controls as well, such as cruise control, taking over the steering wheel, and automated route planning based on real-time traffic information—all aimed at achieving the best-possible fuel efficiency. Perrone’s experience with the DARPA Challenges will once again be key.

As Young has stated in many interviews about the project, he’s not afraid to fail—and he hopes that the LincVolt team can show auto manufacturers a new way of designing cars.

“LincVolt is a great showcase for sustainable transportation,” Perrone says. “If this is something that auto manufacturers could leverage, that would be great.”

For More Information
Java
Oracle Solutions for the Automotive Industry

 


Margaret Harrist is based in Austin, Texas.

 

 
Oracle 1-800-633-0738
    E-mail this page E-mail this page    Printer View Printer View