Compacting asphalt pavement sufficiently is crucial to extending the life of roadways. Intelligent compaction technologies offer methods of evaluating the state of pavement compaction in real time, helping operators to avoid under- or overcompacting the pavement. Most intelligent compaction methods rely on measuring the relative stiffness of the pavement and displaying a real-time, GPS-enabled map of how many passes have been made over the pavement mat.
The Federal Highway Administration has made intelligent compaction a priority investment for the future, and state transportation departments have run numerous tests and demonstration projects on different methods. "Most jobs still don't require the use of intelligent compaction, but they do have pavement density targets," says Fares Beainy, Volvo Construction Equipment research engineer. "So, even then, intelligent compaction can keep you from being penalized."
According to Beainy, the problem is that the density of pavement can be estimated only during paving and confirmed afterward by core tests. "That's the tricky part. Specifications from agencies are in density, but you could only measure the stiffness in real time. But it's density that contractors understand," he says.
Beainy is one of the developers of Volvo Density Direct, an intelligent compaction system that takes a unique approach to calculating pavement density in real time. "The system consists of a GPS antenna, temperature sensors on the front and back of the machine, a touch display and an accelerometer on the drum," says Beainy. "The accelerometer measures the vibration of the drum—other companies use this to estimate the stiffness of the pavement. The difference with Volvo is that we take the same signal from the accelerometer to estimate the density."
Density Direct works by training an onboard computer to recognize the characteristics of a predefined, properly compacted area of the pavement mat; then, it compares measurements on the rest of the paving job to this calibrated, "correct" pavement. All this data is communicated to the operator via a touch screen that shows a color-coded pass map of how close the pavement's characteristics are to the calibrated section. According to Volvo, this can result in pavement densities within 1.5% of pavement core samples—a level of quality control that reduces the need for constant spot tests.
The development of what became Density Direct predates Volvo's involvement. Sesh Commuri, professor of electrical engineering at the University of Oklahoma and one of the inventors of Density Direct, says it was a long road to the finish. "As someone who has worked on this technology for over 13 years, this is a very proud moment for me. Our research has finally seen the light of day."
Commuri's system works with the same temperature sensors and accelerometer data as other intelligent compaction solutions, but the difference is in the way the computer can improve its own performance. "At the heart of our system is an artificial neural network that can 'journalize' and learn as the machine is doing the compaction." The system is able to respond to changing conditions, adapting its model of proper compaction as it gains new data. "It's similar to how a human operator recognizes the performance of a car. If you've driven a car for several years, you can tell, based on the engine noise and vibration, whether you're on a gravel road or a smooth, paved road or if you're on a slight incline," he says. "You take this feedback and use it to calibrate your driving. We are doing the same thing with an artificial neural network in our system."
"Density Direct is a game-changer for us," says Gören Lindgren, president of Volvo CE North America. According to Volvo CE, Density Direct will be available by the end of 2015. Pricing has not yet been announced, but Volvo's intelligent compaction system will be available on three models of rollers in two options: a standard option offering pass mapping and basic intelligent compaction options, and a premium version with the full Density Direct technology.
