Amy J. Briggs

Mobile robot navigation using self-similar landmarks

We propose a new system for vision-based mobile robot navigation in an unmodeled environment. Simple, unobtrusive artificial landmarks are used as navigation and localization aids. The landmark patterns are designed so that they can be reliably detected in real-time in images taken with the robots camera over a wide range of viewing configurations. The code for the recognition algorithm is available in the web site.

Real-time recognition of self-similar landmarks

Abstract Mobile robot localization and navigation require robust and efficient methods for utilizing perceptual information. In this paper we propose a design for simple visual landmarks that can be unobtrusively added to an environment for navigation without a global geometric map of the workspace. We describe self-similar intensity patterns together with an efficient and reliable algorithm for their detection in real-time that can handle a wide range of affine transformations. An implementation of the method is available at www.middlebury.edu/~schar/landmark .

The CS principles project

➧1 In fall 2008 a group of computer scientists and educators convened in Atlanta under the auspices of an NSF-funded conference, Computational Thinking and Fluency in the 21st Century, to discuss the future of computer science education and to identify emerging models. It was widely agreed that students require increasing skills in computing across all disciplines and that a new high school course on a national scale would be an important step towards developing and providing access to these skills. The CS Principles Project has since developed as a collaborative effort involving computer science educators and the College Board with support from the National Science Foundation. Professional organizations including the Computer Science Teachers Association (CSTA) and the Association for Computing Machinery (ACM) have played prominent roles as well in supporting the development of the course and its place within the high school CS curriculum (see Reforming K-12 Computer Science Education ... What Will Your Story Be?, this issue) and the larger CS 10K project (see Transforming High School Computing: A Call to Action, this issue). Support for the project from the computer science education community has led the College Board to commit to making the CS Principles course a new Advanced Placement offering in the comA New Introductory Computing Course for Everyone

An efficient algorithm for one-step planar complaint motion planning with uncertainty

Uncertainty in the execution of robot motion plans must be accounted for in the geometric computations from which plans are obtained, especially in the case where position sensing is inaccurate. We give an &Ogr;(n2 log n) algorithm to find a single commanded motion direction which will guarantee a successful motion in the plane from a specified start to a specified goal whenever such a one-step motion is possible. The plans account for uncertainty in the start position and in robot control, and anticipate that the robot may stick on or slide along obstacle surfaces with which it comes in contact. This bound improves on the best previous bound by a quadratic factor, and is achieved in part by a new analysis of the geometric complexity of the backprojection of the goal as a function of commanded motion direction.

Expected Shortest Paths for Landmark-Based Robot Navigation

In this paper we address the problem of planning reliable landmarkbased robot navigation strategies in the presence of significant sensor uncertainty. The navigation environments are modeled with directed weighted graphs in which edges can be traversed with given probabilities. To construct robust and efficient navigation plans, we compute “expected shortest paths” in such graphs. We formulate the expected shortest paths problem as a Markov decision process and provide two algorithms for its solution. We demonstrate the practicality of our approach using an extensive experimental analysis using graphs with varying sizes and parameters.

Multiple Plane Detection in Image Pairs Using J-Linkage

We present a new method for the robust detection and matching of multiple planes in pairs of images. Such planes can serve as stable landmarks for vision-based urban navigation. Our approach starts from SIFT matches and generates multiple local homography hypotheses using the recent J-linkage technique by Toldo and Fusiello, a robust randomized multi-model estimation algorithm. These hypotheses are then globally merged, spatially analyzed, robustly fitted, and checked for stability. When tested on more than 30,000 image pairs taken from panoramic views of a college campus, our method yields no false positives and recovers 72% of the matchable building walls identified by a human, despite significant occlusions and viewpoint changes.

Offset-polygon annulus placement problems

The @d-annulus of a polygon P is the closed region containing all points in the plane at distance at most @d from the boundary of P. An inner (resp., outer) @d-offset polygon is the polygon defined by the inner (resp., outer) boundary of its @d-annulus. In this paper we address three major problems of covering a given point set S by an offset version or a polygonal annulus of a polygon P. First, the Maximum Cover objective is, given a value of @d, to cover as many points from S as possible by the @d-offset (or by the @d-annulus) of P, allowing translation and rotation. Second, the Containment problem is to minimize the value of @d such that there is a rigid transformation of the @d-offset (or the @d-annulus) of P that covers all points from S. Third, in the Partial Containment problem we seek the minimum offset of P covering k=<|S| points. These problems arise in many applications where one needs to match a given polygonal figure (a known model) to a set of points (usually, obtained measures). We address several variants of these problems, including convex and simple polygons, as well as polygons with holes and sets of polygons, and obtain algorithms with low-degree polynomial running times in all cases.

Robot navigation using 1D panoramic images

This paper presents a new method for navigation and localization of a mobile robot equipped with an omnidirectional camera. We represent the environment using a collection of one-dimensional panoramic images formed by averaging the center scanlines of a cylindrical view. Such 1D images can be stored and processed with few resources, allowing a fairly dense sampling of the environment. Image matching proceeds in real time using dynamic programming on scale-invariant features extracted from each circular view. By analyzing the shape of the matching curve, the relative orientation of pairs of views can be recovered and utilized for navigation. When navigating, the robot continually matches its current view against stored reference views taken from known locations, and determines its location and heading from the properties of the matching results. Experiments show that our method is robust to occlusion, repeating patterns, and lighting variations

An efficient algorithm for one-step planar compliant motion planning with uncertainty

Uncertainty in the execution of robot motion plans must be accounted for in the geometric computations from which plans are obtained, especially in the case where position sensing is inaccurate. We give anO(n2 logn) algorithm to find a single commanded motion direction that will guarantee a successful motion in the plane from a specified start to a specified goal whenever such a one-step motion is possible. The plans account for uncertainty in the start position and in robot control, and anticipate that the robot may stick on or slide along obstacle surfaces with which it comes in contact. This bound improves on the best previous bound by a quadratic factor, and is achieved in part by a new analysis of the geometric complexity of the backprojection of the goal as a function of commanded motion direction.

Matching scale-space features in 1D panoramas

We define a family of novel interest operators for extracting features from one-dimensional panoramic images for use in mobile robot navigation. Feature detection proceeds by applying local interest operators in the scale space of a 1D circular image formed by averaging the center scanlines of a cylindrical panorama. We demonstrate that many such features remain stable over changes in viewpoint and in the presence of noise and camera vibration, and define a feature descriptor that collects shape properties of the scale-space surface and color information from the original images. We then present a novel dynamic programming method to establish globally optimal correspondences between features in images taken from different viewpoints. Our method can handle arbitrary rotations and large numbers of missing features. It is also robust to significant changes in lighting conditions and viewing angle, and in the presence of some occlusion.