Baseball’s trendsetting independent organization, the Atlantic League, may begin using the ball-tracking technology Trackman to call balls and strikes in its games. This automation would regulate the “strike zone” via robotics instead of umpires’ discretion, likely causing the MLB to follow suit. But why do we use robotics?
A recent article published in Baseball America describes sweeping changes facing America’s pastime. One change involves moving the pitcher’s mound further back than the regulation 60 feet and six inches. Another change includes using ball-tracking software to call balls and strikes. This automation can be done with a technological breakthrough known as Trackman that analyzes body movement and ball movement to help golfers and baseball players improve their games. With robotics innovations, we likely wonder: How did we get here? How much has robotics already integrated itself into our lives?
How Robots Fill Needs
The umpire calls a strike or a ball. The players get angry. The crowd boos and jeers. Dust and fists fly. We’ve all seen it a thousand times, haven’t we? In fact, it’s so common that we’ve seen it in cartoons and sitcoms. Until recent years, we’ve depended on the umpire’s judgment—in other words, human perception—to determine the fairness of a baseball in play. However, the game itself is subject to the limitations of the umpire’s eyesight, his brain’s capacity to analyze a 95-mph fastball arcing along a 60-foot pathway, how he extrapolates from incomplete data if he so much as blinks, and his personal biases. In many ways, the “strike zone” serves a perfect example of a need that can be filled by robotics. It’s the need for comprehensive, unbiased, objective data analysis at high speeds.
“To design a robot to do a job, the key is to first understand the nature of the job,” said Dr. John Long, Professor of Biology and Professor of Cognitive Science on the John Guy Vassar Chair of Natural History at Vassar College. “Understanding the job involves understanding two parts: the workplace and the task itself.” In other words, robots need to be aware of the physical area in which they work and of how to perform their programmed tasks.
For example, a mobile vacuum robot like a Roomba must be aware of physical obstacles in the workplace like walls, desks, and chairs. Also, it must have a sensor that determines when the floor is cleaned and then it can move on to a different area. When it runs low on batteries, the robot must combine the two tasks of being aware of the physical area and of how to perform programmed tasks. It can then stop vacuuming and find and approach its charging station. Similarly, Trackman must maintain a 3-D awareness of both the pitcher’s mound and home plate to call a strike or ball.
Complications Implementing Robotics
“When we look at a workplace, a big deal in robotics is to think about how stable it is,” Dr. Long said. “A stable workplace we call a structured workplace, and it’s one in which the world doesn’t change, or doesn’t change much.” Few things move or change in a structured, stable work environment. Dr. Long offers a factory floor with an assembly line as an example of a highly structured and stable workplace.
On the other hand, less static workplaces are referred to as unstructured workplaces. “Unstructured workplaces change all the time and they change in ways that are not often predictable,” Dr. Long said. “People, animals, vehicles come and they go; rain falls.”
A baseball diamond falls in the middle of that spectrum. Its physical structure stays the same. The distance from the pitcher’s mound to home plate remains fixed. Even the general placement of the players and umpires falls within general parameters. However, a league-wide, unbiased tool like Trackman must account for variables like weather, visual environment changes from stadium to stadium, camera angles, determining follow-through on batters’ swings, and so on.
As humans delegate increasingly complicated tasks to robots, we must plan for highly structured tasks that can be automated due to workplace stability in terms of the work space. These initial requirements of stabilizing task execution and the workplace environment serve as just two examples of the road ahead for robotic automation.
Dr. John Long contributed to this article.
Dr. Long is a Professor of Biology and a Professor of Cognitive Science on the John Guy Vassar Chair of Natural History at Vassar College. He also serves as the Director of Vassar’s Interdisciplinary Robotics Research Laboratory, which he helped found in 2003. Professor Long received his Ph.D. in Zoology from Duke University.