dstrelow,ssingh¢ Abstract — Position estimation for planetary rovers has been typically limited to odometry based on proprioceptive measurements such as the integration of distance traveled and measurement of heading change. Here we present and compare two methods of online visual odometry suited for planetary rovers. Both methods use omnidirectional imagery to estimate motion of the rover. One method is based on robust estimation of optical flow and subsequent integration of the flow. The second method is a full structure-from-motion solution. To make the comparison meaningful we use the same set of raw corresponding visual features for each method. The dataset is an sequence of 2000 images taken during a field experiment in the Atacama desert, for which high resolution GPS ground truth is available. I.

Abstract — This paper describes a novel experiment in which two very different methods of underwater robot localization are compared. The first method is based on a geometric approach in which a mobile node moves within a field of static nodes, and all nodes are capable of estimating the range to their neighbours acoustically. The second method uses visual odometry, from stereo cameras, by integrating scaled optical flow. The fundamental algorithmic principles of each localization technique is described. We also present experimental results comparing acoustic localization with GPS for surface operation, and a comparison of acoustic and visual methods for underwater operation. I.

Abstract not found

Abstract — Dynamic legged locomotion entails navigating unstructured terrain at high speed. The discontinuous foot-fall patterns and flight phases, which are pivotal for its unrivaled mobility, introduce large impulses and extended free-falls that serve to destabilize motion estimation. In a nod to biological systems, visual information, in the form of optical flow, is used with a hybrid estimator tuned to the principal phases of legged locomotion. This takes advantage of the ballistic nature of the flight phases to vary optical flow calculation methods and estimator parameters. Experimentation on a single-leg shows a reduction in inertial drift. In tests with 6g impulses, pose was recovered within 5 deg rms with angular rate errors limited to 10 deg/sec at frequencies up to 250 Hz. This compares well with angular rate recovery by vision only and traditional inertial techniques. I.

Abstract — We describe a multi-sensory control architecture for hovering of a mini four-rotor unmanned aerial vehicle (UAV) over specified markers as an external reference for pose estimation utilizing two on-board sensors: a tiny single camera and Inertial Measurement Units. A high-speed pose estimation approach based on visual information is presented. A closed-loop system is implemented using PID pose-controllers. The control performance is improved by integration of IMU measurement. Results of a real-time experiment are presented. Index Terms — Vision system, pose estimation, multiple sensing system, UAV/MAV control. I.

Abstract — This paper describes a novel experiment in which two very different methods of underwater robot localization are compared. The first method is based on a geometric approach in which a mobile node moves within a field of static nodes, and all nodes are capable of estimating the range to their neighbours acoustically. The second method uses visual odometry, from stereo cameras, by integrating scaled optical flow. The fundamental algorithmic principles of each localization technique is described. We also present experimental results comparing acoustic localization with GPS for surface operation, and a comparison of acoustic and visual methods for underwater operation. I.

Autonomously operating unmanned aerial vehicles (UAV) have a great potential for many applications such as reconnaissance, mapping and surveillance. Whilst needing low cost and light weight navigation systems in their implementations, sensors like GPS or low cost inertial sensors can’t separately provide either the complete set of information or the required degree of accuracy. Multi-sensor solutions have to be sought to mitigate the shortcomings of individual sensors. Often in such systems, low cost MEMS INS is aided by GPS to provided position, velocity and attitude (PVA) measurements. Data fusion strategies and techniques like extended Kalman filter (EKF) are the kernel for estimating and compensating individual sensor errors to reach improved performance. One drawback of integrating GPS with low cost MEMS INS is the heavy reliance on GPS signal availability. The accuracy of PVA solutions would degrade sharply during GPS signal

Abstract — An air-ground multi-robot system is designed for the purpose of applying bio-inspired sensor-motor modeling on technical systems. It consists of a flying mini-quadrotor equipped with inertial and visual sensors, and a wheeled minirobot equipped with active markers. The system modules, a multi-sensory pose/motion estimation approach and a closedloop control of the flying quadrotor are described in this paper. The quadrotor can hover over the ground robot stably and track the ground robot movement. An accurate pose estimation is verified by a comparison between the estimated quadrotor pose and the actual pose during the flight. I.

Abstract—This paper is concerned with the problem of autonomously landing an unmanned aerial vehicle (UAV) on a stationary platform. Our solution consists of two parts, a sensor fusion framework producing estimates of the UAV state and a control system that computes appropriate actuator commands. There are three sensors used, a camera, a GPS and a compass. Besides the description of the solution, we also present experimental results illustrating the results obtained in using our system to autonomously land an UAV. I.

Abstract not found