One of the goals in the field of mobile robotics is the development of mobile platforms which operate in populated environments. For many tasks it is therefore highly desirable that a robot can track the positions of the humans in its surrounding. In this paper we introduce sample-based joint probabilistic data association filters as a new algorithm to track multiple moving objects. Our method applies Bayesian filtering to adapt the tracking process to the number of objects in the perceptual range of the robot. The approach has been implemented and tested on a real robot using laser-range data. We present experiments illustrating that our algorithm is able to robustly keep track of multiple persons. The experiments furthermore show that the approach outperforms other techniques developed so far.

Over the last years, particle filters have been applied with great success to a variety of state estimation problems. In this paper we present a statistical approach to increasing the efficiency of particle filters by adapting the size of sample sets during the estimation process.

Estimating the location of people using a network of sensors placed throughout an environment is a fundamental challenge in ubiquitous computing and smart environments. Id-sensors such as infrared badges provide explicit object identity information but coarse location information while anonymous sensors such as laser range-finders provide accurate location information only. Tracking using both sensor types simultaneously is an open research challenge. We present a novel approach to tracking multiple objects that combines the accuracy benefits of anonymous sensors and the identification certainty of id-sensors. Rao-Blackwellised particle filters are used to estimate object locations. Each particle represents the association history between Kalman filtered object tracks and observations. After using only anonymous sensors until id estimates are certain enough, id assignments are sampled as well resulting in a fully Rao-Blackwellised particle filter over both object tracks and id assignments. Our approach was implemented and tested successfully using data collected in an indoor environment.

Abstract. Location estimation is an important part of many ubiquitous computing systems. Particle filters are simulation-based probabilistic approximations which the robotics community has shown to be effective for tracking robots ’ positions. This paper presents a case study of applying particle filters to location estimation for ubiquitous computing. Using trace logs from a deployed multi-sensor location system, we show that particle filters can be as accurate as common deterministic algorithms. We also present performance results showing it is practical to run particle filters on devices ranging from high-end servers to handhelds. Finally, we discuss the general advantages of using probabilistic methods in location systems for ubiquitous computing, including the ability to fuse data from different sensor types and to provide probability distributions to higher-level services and applications. Based on this case study, we conclude that particle filters are a good choice to implement location estimation for ubiquitous computing. 1

Probabilistic inference is the problem of estimating the hidden states of a system in an optimal and consistent fashion given a set of noisy or incomplete observations. The optimal solution to this problem is given by the recursive Bayesian estimation algorithm which recursively updates the posterior density of the system state as new observations arrive online.

In this paper, we propose a way to fuse the image-based localisation approach with the Monte-Carlo localisation approach. The method we propose does not suffer of the major limitation of the two separated methods: the need of a metric map of the environment for the Monte-Carlo localisation and the failure of the image-based approach in environments with spatial periodicity (perceptual aliasing). The approach we developed exploits the properties of the Fourier Transform of the omnidirectional images and uses the similarity between the images to weights the beliefs about the robot position. Successful experiments in large indoor environment are presented in which we do not used a priory information on the metrical map of the environment.

We propose a novel approach to optimize Partially Observable Markov Decisions Processes (POMDPs) defined on continuous spaces. To date, most algorithms for model-based POMDPs are restricted to discrete states, actions, and observations, but many real-world problems such as, for instance, robot navigation, are naturally defined on continuous spaces. In this work, we demonstrate that the value function for continuous POMDPs is convex in the beliefs over continuous state spaces, and piecewise-linear convex for the particular case of discrete observations and actions but still continuous states. We also demonstrate that continuous Bellman backups are contracting and isotonic ensuring the monotonic convergence of value-iteration algorithms. Relying on those properties, we extend the PERSEUS algorithm, originally developed for discrete POMDPs, to work in continuous state spaces by representing the observation, transition, and reward models using Gaussian mixtures, and the beliefs using Gaussian mixtures or particle sets. With these representations, the integrals that appear in the Bellman backup can be computed in closed form and, therefore, the algorithm is computationally feasible. Finally, we further extend PERSEUS to deal with continuous action and observation sets by designing effective sampling approaches.

Abstract — Particle filters have recently been applied with great success to mobile robot localization. This success is mostly due to their simplicity and their ability to represent arbitrary, multi-modal densities over a robot’s state space. The increased representational power, however, comes at the cost of higher computational complexity. In this paper we introduce adaptive real-time particle filters that greatly increase the performance of particle filters under limited computational resources. Our approach improves the efficiency of state estimation by adapting the size of sample sets on-the-fly. Furthermore, even when large sample sets are needed to represent a robot’s uncertainty, the approach takes every sensor measurement into account, thereby avoiding the risk of losing valuable sensor information during the update of the filter. We demonstrate empirically that this new algorithm drastically improves the performance of particle filters for robot localization. I.

Location information is an important source of context for ubiquitous computing systems. This paper looks at how a foot-mounted inertial unit, a detailed building model, and a particle filter can be combined to provide absolute positioning, despite the presence of drift in the inertial unit and without knowledge of the user’s initial location. We show how to handle multiple floors and stairways, how to handle symmetry in the environment, and how to initialise the localisation algorithm using WiFi signal strength to reduce initial complexity. We evaluate the entire system experimentally, using an independent tracking system for ground truth. Our results show that we can track a user throughout a 8725 m 2 building spanning three floors to within 0.5 m 75 % of the time, and to within 0.73 m 95 % of the time. ACM Classification Keywords

this report. Kautz [79] modeled plan recognition logically in a manner that allowed goals and plans to be described at various levels of abstraction. Etzioni et al. [94, 95, 92, 93] developed a version space algorithm for plan recognition that is provably sound and polynomial time [94, 93]. Weld et al. developed goal recognition algorithms using inductive logic programming [90] and version-space algebra [89, 168, 88] in the context of programming by demonstration