This paper demonstrates that computer vision techniques can estimate the heading of a small fixed pitch four rotor helicopter. Heading estimates are computed using the optical flow technique of phase correlation on images captured using a down facing camera. The camera is fitted with an omnidirectional lens and the images are transformed into the logpolar domain before the main computational step. The vision algorithm runs at 10 Hz on a single board computer (SBC) mounted aboard the craft. Experimental performance of this system is compared with results obtained from a traditional inertial measurement unit (IMU). It is found that the yaw rate computed from the optical flow is comparable to the IMU and thus appropriate for use in controlling the helicopter.
INTRODUCTIONRobot control and navigation require motion and attitude sensing mechanisms. Tasks such as object avoidance, path planning and navigation all benefit from ego-motion estimation and motion estimation of the environment. A robust estimator of robot state must provide timely and accurate information in order for a craft to make the necessary control and navigation decisions. Such a system is typically based on a combination of sensors and an estimator which fuses the data into a more accurate estimate.In this paper we are interested in controlling a small fixed pitch four rotor helicopter called a quadrotor, Figure 1. Unpiloted aerial vehicles, like this, are usually controlled using an attitude estimate obtained from an inertial measurement unit (IMU) consisting of gyroscopes, magnetometers, accelerometers and barometer. However, we are interested in assessing the performance of a vision system in estimating the heading. Our interest in vision is to overcome problems in IMUs, such as drift, which accumulates, and problems in GPS with loss of signal for prolonged periods of time. If it can be shown that vision systems are capable of similar performance then integrating them into the flight control system seems sensible.A vision system must not only be capable of providing attitude estimates, but be able to provide these in a timely manner to have effective control, and on hardware small and light enough to be mounted on the vehicle. In this paper we will be looking at such a practical system that employs two separate computer systems, as can be seen in Figure 2. The flight control system is an ARM7 microprocessor based system that includes inputs from the IMU. A separate vision processing system, including camera and lens, is implemented on a Gumstix Overo single board computer (SBC). The vision system communicates to the flight control via a USB interface.For the purposes of this comparative work we shall limit ourselves to measuring only the components of yaw 1 rate, r in the aircraft body-frame, although the calculations also yield altitude Z, from a downward facing camera fitted with a fish eye lens. We shall present the paper in the following order. Firstly, a discussion of the optical flow process for estimating motion from imagery is gi...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.