Vision-Based Autonomous Landing of an Helicopter on a Moving Target

This paper presents the design and implementation of an algorithm for landing a helicopter autonomously on a moving target. The target is neutral; it is neither adversarial nor cooperative. A downward-looking camera is used to acquire and track the target. A template matching algorithm is used for acquiring the target in the images. This is integrated with a trajectory controller for landing the helicopter. The position of the target in the image is the input to a robust Kalman lter for tracking in successive frames. A linear controller based on a kinematic model of the helicopter is used to perform trajectory following and landing. We present results from experimental trials showing successful trajectory tracking and landing on moving targets.