NASA/JPL's Mars exploration rovers acquire their attitude upon command and autonomously propagate their attitude and position. The rovers use accelerometers and images of the sun to acquire attitude, autonomously searching the sky for the sun with an articulated camera. To propagate the attitude and position the rovers use either accelerometer and gyro readings or gyro readings and wheel odometry, depending on the nature of the movement Earth-based operators have commanded. Where necessary, visual odometry is performed on images to fine tune the position updates, particularly in high slip environments. The capability also exists for visual odometry attitude updates. This paper describes the techniques used by the rovers to acquire and maintain attitude and position knowledge, the accuracy which is obtainable, and lessons learned after more than one year in operation.

On January 24, 2004, the Mars Exploration Rover named Opportunity successfully landed in the region of Mars known as Meridiani Planum, a vast plain dotted with craters where orbiting spacecraft had detected the signatures of minerals believed to have formed in liquid water. The first pictures back from Opportunity revealed that the rover had landed in a crater roughly 20 meters in diameter - the only sizeable crater within hundreds of meters - which became known as Eagle Crater. And in the walls of this crater just meters away was the bedrock MER scientists had been hoping to find, which would ultimately prove that this region of Mars did indeed have a watery past. Opportunity explored Eagle Crater for almost two months, then drove more than 700 meters in one month to its next destination, the much larger Endurance Crater. After surveying the outside of Endurance Crater, Opportunity drove into the crater and meticulously studied it for six months. Then it went to examine the heat shield that had protected Opportunity during its descent through the Martian atmosphere. More than a year since landing, Opportunity is still going strong and is currently en route to Victoria Crater - more than six kilometers from Endurance Crater. Opportunity has driven more than four kilometers, examined more than eighty patches of rock and soil with instruments on the robotic arm, excavated four trenches for subsurface sampling, and sent back well over thirty thousand images of Mars - ranging from grand panoramas to up close microscopic views. This paper details the experience of driving Opportunity through this alien landscape from the point of view of the Rover Planners, the people who tell the rover where to drive and how to use its robotic arm.

NASA's Mars Exploration Rovers (MER) was designed to traverse in Viking Lander-I style terrains: mostly flat, with many small non-obstacle rocks and occasional obstacles. During actual operations in such terrains, onboard position estimates derived solely from the onboard inertial measurement unit and wheel encoder-based odometry achieved well within the design goal of at most 10% error. However, MER vehicles were also driven along slippery slopes tilted as high as 31 degrees. In such conditions an additional capability was employed to maintain a sufficiently accurate onboard position estimate: visual odometry. The MER visual odometry system comprises onboard software for comparing stereo pairs taken by the pointable mast-mounted 45 degree FOV navigation cameras (NAV-CAMs). The system computes an update to the 6-DOF rover pose (x, y, z, roll, pitch, yaw) by tracking the motion of autonomously-selected "interesting" terrain features between two pairs of stereo images, in both 2D pixel and 3D world coordinates. A maximum likelihood estimator is applied to the computed 3D offsets to produce a final, corrected estimate of vehicle motion between the two pairs. In this paper we describe the visual odometry algorithm used on the Mars Exploration Rovers, and summarize its results from the first year of operations on Mars.

Spirit is one of two rovers that landed on Mars in January 2004 as part of NASA's Mars Exploration Rover mission. As of July 2005, Spirit has traveled over 4.5 kilometers across the Martian surface while investigating rocks and soils, digging trenches to examine subsurface materials, and climbing hills to reach outcrops of bedrock. Originally designed to last 90 sols (Martian days), Spirit has survived over 500 sols of operation and continues to explore. During the mission, we achieved increases in efficiency, accuracy, and traverse capability through increasingly complex command sequences, growing experience, and updates to the on-board and ground-based software. Safe and precise mobility on slopes and in the presence of obstacles has been a primary factor in development of new software and techniques.

Increased attention has been focused in recent years on human-machine systems, how they are architected, and how they should operate. The purpose of this paper is to describe an actual instance of a practical human-robot system used on a NASA Mars rover mission that has been underway since January 2004 involving daily interaction between humans on Earth and mobile robots on Mars. The emphasis is on the human-robot collaborative arrangement and the performance enabled by mobility and robotic arm software functionality during the first 90 days of the mission. Mobile traverse distance, accuracy, and rate as well as robotic arm operational accuracy achieved by the system is presented.

Robust navigation for mobile robots over long distances requires an accurate method for tracking the robot position in the environment. Techniques for position estimation by determining the camera ego-motion from monocular or stereo sequences have been previously described. However, long-distance navigation requires a very high level of robustness and a very low rate of error growth. In this paper, we describe a methodology for long-distance rover navigation that meets these goals using robust estimation. We show that a system based on only camera ego-motion estimates will accumulate errors with super-linear growth in the distance travelled, owing to increasing orientation errors. When an absolute orientation sensor is incorporated, the error growth can be reduced to a linear function of the distance travelled. We tested these techniques using both extensive simulation and hundreds of real rover images and achieved a low, linear rate of error growth.

Several methods for computing observer motion from monocular and stereo image sequences have been proposed. However, accurate positioning over long distances requires a higher level of robustness than previously achieved. This paper describes several mechanisms for improving robustness in the context of a maximum-likelihood stereo ego-motion method. We demonstrate that even a robust system will accumulate super-linear error in the distance traveled due to increasing orientation errors. However, when an absolute orientation sensor is incorporated, the growth is reduced to linear in the distance traveled, grows much more slowly in practice. Our experiments, including a trial with 210 stereo pairs, indicate that these techniques can achieve errors below 1% of the distance traveled. This method has been implemented to run on-board a prototype Mars rover.

Manipulators mounted on-board rovers have limited dexterity due to power and weight constraints imposed by rover designs. However, to perform science operations, it is necessary to be able to position and orient these manipulators on science targets in order to carry out in-situ measurements. This article describes how we enhance manipulator dexterity using the rover mobility system. The lack of omni-directional driving capability and the constraints imposed by the mobility mechanism requires vehicle maneuvering to supplement the manipulators' motions. Target tracking using stereo vision is integrated with rover maneuvering to perform two types of operations: rock sample acquisition for return to earth and instrument placement for in-situ science measurements. We describe the computational architecture, tools, and algorithms that we developed for this task. We have successfully demonstrated these operations on a self-contained Mars Rover prototype, Rocky 7. We have demonstrated grasping a small rock sample from a distance of more than one meter away and placing an instrument on a boulder from a distance of more than five meters away.

In June and July 1997, Nomad, a planetary-relevant mobile robot, traversed more than 220 kilometers across the barren Atacama Desert in Chile, exploring a landscape analogous to the surfaces of the Moon and Mars. In this unprecedented demonstration, Nomad operated both autonomously and under the control of operators thousands of kilometers away, addressing issues of robot configuration, communication, position estimation, and navigation in rugged, natural terrain. The field experiment also served to test technologies for remote geological investigation, paving the way for new exploration strategies on Earth and beyond. Finally, by combining safeguarded teleoperation with panoramic visualization and a novel user interface, the Atacama Desert Trek provided the general public a compelling interactive experience an opportunity to remotely drive an exploratory robot. Nomad's performance in the Atacama Desert Trek set new benchmark in high performance robotics operations relevant to terrestrial and planetary exploration. This paper presents an overview of the experiment, describes technologies key to Nomad's success, and discusses outcomes and implications.

We present a stereoscopic mapping system for use in post-nuclear accident operations by the Pioneer robot. First we discuss a radiation shielded sensor array designed to tolerate extended cumulative dose using 4/spl times/ shielding. Next, we outline procedures to ensure timely, accurate range estimation using trinocular stereo. Finally, we review the implementation of a system for the integration of range information into a 3-D, textured, metrically accurate surface mesh.

Many aspects of the real world continue to plague stereo matching systems. One of these is perspective foreshortening, an effect that occurs when a surface is viewed at a sharp angle. Because each stereo camera has a slightly different view, the image of the surface is more compressed and occupies a smaller area in one view. These effects cause problems because most stereo methods compare similarly sized regions (using the same-sized windows in both images), tacitly assuming that objects occupy the same extents in both images. Clearly this condition is violated by perspective foreshortening. We show how to overcome this problem using a local spatial frequency representation. A simple geometric analysis leads to an elegant solution in the frequency domain which, when applied to a Gabor filter-based stereo system, increases the system's maximum matchable surface angle from 30 degrees to over 75 degrees.

Miro is a set of languages and tools that support the visual specification of file system security. Two visual languages are presented: the instance language, which allows specification of file system access, and the constraint language, which allows specification of security policies. Miro visual languages and tools are used to specify security configurations. A visual language is one whose entities are graphical, such as boxes and arrows, specifying means stating independently of any implementation the desired properties of a system. Security means file system protection: ensuring that files are protected from unauthorized access and granting privileges to some users, but not others. Tools implemented and examples of how these languages can be applied to real security specification problems are described.>