Andrew M. Ladd

Practical robust localization over large-scale 802.11 wireless networks

We demonstrate a system built using probabilistic techniques that allows for remarkably accurate localization across our entire office building using nothing more than the built-in signal intensity meter supplied by standard 802.11 cards. While prior systems have required significant investments of human labor to build a detailed signal map, we can train our system by spending less than one minute per office or region, walking around with a laptop and recording the observed signal intensities of our buildings unmodified base stations. We actually collected over two minutes of data per office or region, about 28 man-hours of effort. Using less than half of this data to train the localizer, we can localize a user to the precise, correct location in over 95% of our attempts, across the entire building. Even in the most pathological cases, we almost never localize a user any more distant than to the neighboring office. A user can obtain this level of accuracy with only two or three signal intensity measurements, allowing for a high frame rate of localization results. Furthermore, with a brief calibration period, our system can be adapted to work with previously unknown user hardware. We present results demonstrating the robustness of our system against a variety of untrained time-varying phenomena, including the presence or absence of people in the building across the day. Our system is sufficiently robust to enable a variety of location-aware applications without requiring special-purpose hardware or complicated training and calibration procedures.

Robotics-based location sensing using wireless ethernet

A key subproblem in the construction of location-aware systems is the determination of the position of a mobile device. This paper describes the design, implementation and analysis of a system for determining position inside a building from measured RF signal strengths of packets on an IEEE 802.11b wireless Ethernet network. Previous approaches to location awareness with RF signals have been severely hampered by non-linearity, noise and complex correlations due to multi-path effects, interference and absorption. The design of our system begins with the observation that determining position from complex, noisy and non-linear signals is a well-studied problem in the field of robotics. Using only off-the-shelf hardware, we achieve robust position estimation to within a meter in our experimental context and after adequate training of our system. We can also coarsely determine our orientation and can track our position as we move. By applying recent advances in probabilistic inference of position and sensor fusion from noisy signals, we show that the RF emissions from base stations as measured by off-the-shelf wireless Ethernet cards are sufficiently rich in information to permit a mobile device to reliably track its location.

On the feasibility of using wireless ethernet for indoor localization

IEEE 802.11b wireless Ethernet is becoming the standard for indoor wireless communication. This paper proposes the use of measured signal strength of Ethernet packets as a sensor for a localization system. We demonstrate that off-the-shelf hardware can accurately be used for location sensing and real-time tracking by applying a Bayesian localization framework.

Wireless LAN location-sensing for security applications

This paper considers the problem of using wireless LAN location-sensing for security applications. Recently, Bayesian methods have been successfully used to determine location from wireless LAN signals, but such methods have the drawback that a model must first be built from training data. The introduction of model error can drastically reduce the robustness of the location estimates and such errors can be actively induced by malicious users intent on hiding their location. This paper provides a technique for increasing robustness in the face of model error and experimentally validates this technique by testing against unmodeled hardware, modulation of power levels, and the placement of devices outside the trained workspace. Our results have interesting ramifications for location privacy in wireless networks.

Robotics-based location sensing using wireless Ethernet

Abstract A key subproblem in the construction of location-aware systems is the determination of the position of a mobile device. This article describes the design, implementation and analysis of a system for determining position inside a building from measured RF signal strengths of packets on an IEEE 802.11b wireless Ethernet network. Previous approaches to location-awareness with RF signals have been severely hampered by non-Gaussian signals, noise, and complex correlations due to multi-path effects, interference and absorption. The design of our system begins with the observation that determining position from complex, noisy and non-Gaussian signals is a well-studied problem in the field of robotics. Using only off-the-shelf hardware, we achieve robust position estimation to within a meter in our experimental context and after adequate training of our system. We can also coarsely determine our orientation and can track our position as we move. Our results show that we can localize a stationary device to within 1.5 meters over 80% of the time and track a moving device to within 1 meter over 50% of the time. Both localization and tracking run in real-time. By applying recent advances in probabilistic inference of position and sensor fusion from noisy signals, we show that the RF emissions from base stations as measured by off-the-shelf wireless Ethernet cards are sufficiently rich in information to permit a mobile device to reliably track its location.

Using wireless Ethernet for localization

IEEE 802.11b wireless Ethernet is rapidly becoming the standard for in-building and short-range wireless communication. Many mobile devices such as mobile robots, laptops and PDAs already use this protocol for wireless communication. Many wireless Ethernet cards measure the signal strength of incoming packets. This paper investigates the feasibility of implementing a localization system using this sensor. Using a Bayesian localization framework, we show experiments demonstrating that off-the-shelf wireless hardware can accurately be used for location sensing and tracking with about one meter precision in a wireless-enabled office building.

Measure theoretic analysis of probabilistic path planning

This paper presents a novel analysis of the probabilistic roadmap method (PRM) for path planning. We formulate the problem in terms of computing the transitive closure of a relation over a probability space, and give a bound on the expected number iterations of PRM required to find a path, in terms of the number of intermediate points and the probability of choosing a point from a certain set. Explicit geometric assumptions are not necessary to complete this analysis. As a result, the analysis provides some unification of previous work. We provide an upper bound which could be refined using details specific to a given problem. This bound is of the same form as that proved in previous analyses, but has simpler prerequisites and is proved on a more general class of problems. Using our framework, we analyze some new path-planning problems, 2k-degree-of-freedom kinodynamic point robots, polygonal robots with contact, and deformable robots with force field control. These examples make explicit use of generality in our approach that did not exist in previous frameworks.

Simulated knot tying

Applications such as suturing in medical simulations require the modeling of knot tying in physically realistic rope. The paper describes the design and implementation of such a system. Our model uses a spline of linear springs, adaptive subdivision and a dynamics simulation. Collisions are discrete event simulated and follow the impulse model. Although some care must be taken to maintain stable knots, we demonstrate our simple model is sufficient for this task. In particular, we do not use friction or explicit constraints to maintain the knot. As examples, we tie an overhand knot and a reef knot.

Motion Planning in the Presence of Drift, Underactuation and Discrete System Changes

Motion planning research has been successful in developing planning algorithms which are effective for solving problems with complicated geometric and kinematic constraints. Various applications in robotics and in other fields demand additional physical realism. Some progress has been made for non-holonomic systems. However systems with significant dr ift, underactuation and discrete system changes remain challenging for existing planning techniques particularly as the dimensional- ity of the state space increases. In this paper, we demonstrate a motion planning technique for the solution of problems with these challenging characteristics. Our approach uses sampling-based motion planning and subdivision methods. The problem that we solve is a game that was chosen to exemplify characteristics of dynamical systems that are difficult for planning. To our knowledge, this is first application of algorithmic motion p lanning to a problem of this type and complexity.

Probabilistic Roadmaps of Trees for Parallel Computation of Multiple Query Roadmaps

We propose the combination of techniques that solve multiple queries for motion planning problems with single query planners in a motion planning framework that can be efficiently parallelized. In multiple query motion planning, a data structure is built during a preprocessing phase in order to quickly respond to on-line queries. Alternatively, in single query planning, there is no preprocessing phase and all computations occur during query resolution. This paper shows how to effectively combine a powerful sample-based method primarily designed for multiple query planning (the Probabilistic Roadmap Method - PRM) with sample-based tree methods that were primarily designed for single query planning (such as Expansive Space Trees, Rapidly Exploring Random Trees, and others). Our planner, which we call the Probabilistic Roadmap of Trees (PRT), uses a tree algorithm as a subroutine for PRM. The nodes of the PRM roadmap are now trees. We take advantage of the very powerful sampling schemes of recent tree planners to populate our roadmaps. The combined sampling scheme is in the spirit of the non-uniform sampling and refinement techniques employed in earlier work on PRM. PRT not only achieves a smooth spectrum between multiple query and single query planning but it combines advantages of both. We present experiments which show that PRT is capable of solving problems that cannot be addressed efficiently with PRM or single-query planners. A key advantage of PRT is that it is significantly more decoupled than PRM and sample-based tree planners. Using this property, we designed and implemented a parallel version of PRT. Our experiments show that PRT distributes well and can easily solve high dimensional problems that exhaust resources available to single machines.