New research suggests how skateboarders behave You can use math to spice up your halfpipe game.
Tricks like kickflips and ollies require speed. Skateboarders gain speed and pump higher as they roll along a U-shaped ramp called a halfpipe. Alternate between squatting and standing to use the pump.
According to Florian Kogelbauer, pumping on a halfpipe is very similar to pumping on a swing. A mathematician, he works at ETH Zurich in Switzerland. On the swing, lean back to stretch your legs and lean forward to bend your knees. This movement changes the way your weight is distributed. Synchronizing the back and forth movement of your swing with the movement of your pump will increase your energy and propel you to greater heights. This increases potential energy. Pumping in a halfpipe does something similar for skateboarders.
Where should skateboarders change positions to get the biggest boost? Some say it's important to practice and feel the rhythm. Kogelbauer and his colleagues turned to mathematics instead.
They used theoretical research. This type starts by taking a real-world problem and converting it into an equation, Kogelbauer says. Then solve these math problems and translate the answers to the real world.
The researchers started with equations for the simple case of a rider moving back and forth on a swing. Most aspects of the swing are well understood: the forces at play, the motion, and the energy involved. The scientists then modified these equations to better fit the skateboarding scenario. They used them to create mathematical models. How far a skater can go up a ramp is determined by how their body position changes along the halfpipe. This model captured that relationship.
They then programmed the simple model into a computer. Added real world details. This includes things like halfpipe size and skateboarder size. The computer model then calculated the skater's ideal body position as he moved along the length of the ramp.
The model suggests that skateboarders should crouch downhill and stand up just before reaching a flat area. After that, you will have to crouch again until you reach the uphill slope. There, they must stand up again and continue standing until they begin to roll downhill.
A simple model of real-world motion
The next step was to compare the model's output to the movements of a real skateboarder. Researchers recruited two skaters to try out the halfpipe. The scientists instructed them to reach a certain ramp height as quickly as possible. The team then analyzed the video footage to see where the skaters were crouching and standing.
More experienced skaters pumped naturally as the team's model suggested and reached goal height faster. Less experienced skateboarders had to pump more times to reach the same height. And the skater's body position didn't match the model very well.
This means the model likely captures the most efficient method of pumping, Kogelbauer concludes. If you want to get better at halfpipe skating, try squatting or standing as the model shows, he says. The researchers shared their results in August: physical review study.
The research could also be useful in robotics, Solina Lupu says. An engineer, she studies ways to help robots walk at the California Institute of Technology in Pasadena. A person can contort his body in various ways. But robots don't have muscles. This limits your range of motion, she explains. This makes it difficult to program complex movements such as walking.
Some groups use high-tech equipment to analyze human movements. It then uses artificial intelligence and machine learning to enable the robot to imitate those movements. While this may be effective, it may complicate the robot's behavioral model. And when something goes wrong, Lupu says it can be difficult to understand what happened and why.
However, skateboarding research used a simple model to predict how a person should move. It worked, even though it didn't include all the complex powers. This approach could also be applied to robotics, she says.
“There's a lot of value in the work they've done,” Lupu said. She particularly likes that the team checked the model's results against the movements of experienced skateboarders.
This agreement shows that simple systems can successfully model real-world behavior, Kogelbauer said. And it suggests that such models could help improve the performance of both athletes and robots.
Source: www.snexplores.org
