Photo Taken Using SLAM Algorithm during Tests at the Great Plains Joint Training Center’s Crisis City
Kansas State University’s Dale Schinstock has developed and continues to refine an algorithm that allows sensors on an unmanned aircraft to build a map of the environment and determine its location in that map.
The Simultaneous Localization and Mapping algorithm is based on a single camera and relies on a specific kind of sensor. “More complicated sensors can’t be used as part of an unmanned aircraft system because they’re too heavy,” explained Schinstock, an associate professor of mechanical and nuclear engineering at K-State. “We had to use a more simplistic sensor, but it had to accomplish the same thing as its larger counterparts that use multiple cameras.”
The larger sensors are often used on ground vehicles, where weight and size are less crucial. The SLAM algorithm, developed based on open source code, tells the unmanned aircraft’s ground crew where the vehicle is at and lets the team hold it in one position.
Schinstock predicts that in the future, the system will be used in disaster situations. “SLAM lets the operator fly the vehicle in urban canyons where tall buildings can affect global positioning systems,” said Schinstock. “That will allow first responders to get a more complete picture of the scene which will better equip them to handle the situation more effectively and efficiently.”
Source: Salina K State Newsletter