Algorithm corrects for video latency, delivers precision control of unmanned ground vehicles
Summary
A Javelin missile was fired from a Milrem Robotics unmanned ground vehicle using a Kongsberg remote launcher at the U.S. Armys Redstone Arsenal in Alabama. According to Dr. Mark Brudnak from the Army’s CCDC Ground Vehicle Systems Center, network latency causes operators to adopt a “move and wait” style of driving, which is too slow to be useful in many situations. In a conference paper (linked below), Brudnak explained how his algorithm splits the video image, transforms the perspective, and stitches it back together to give an accurate point of view to the operator “in the most natural way.” “Preliminary experimental results were presented in which the predictive display was shown to be an effective method for the mitigation of latency by increasing achieved speed and by reducing the path deviation and the heading error significantly,” Brudnak wrote. In this case, a patent license agreement with the Army would enable the algorithm and related tech developed by Brudnak to be transferred and transitioned into use by companies selling products and services to the military or commercial markets. Companies working on self-driving cars, like Designated Driver, might be interested in learning and licensing teleoperation technologies developed by the Army, Metzger said.