Dirk Haehnel

The Mobile Sensing Platform: An Embedded Activity Recognition System

Activity-aware systems have inspired novel user interfaces and new applications in smart environments, surveillance, emergency response, and military missions. Systems that recognize human activities from body-worn sensors can further open the door to a world of healthcare applications, such as fitness monitoring, eldercare support, long-term preventive and chronic care, and cognitive assistance. Wearable systems have the advantage of being with the user continuously. So, for example, a fitness application could use real-time activity information to encourage users to perform opportunistic activities. Furthermore, the general public is more likely to accept such activity recognition systems because they are usually easy to turn off or remove.

Are GSM Phones THE Solution for Localization

In this paper, we argue that localization solution based on cellular phone technology, specifically GSM phones, is a sufficient and attractive option in terms of coverage and accuracy for a wide range of indoor, outdoor, and placebased location-aware applications. We present preliminary results that indicate that GSM-based localization systems have the potential to detect the places that people visit in their everyday lives, and can achieve median localization accuracies of 5 and 75 meters for indoor and outdoor environments, respectively.

Gaussian Processes and Reinforcement Learning for Identification and Control of an Autonomous Blimp

Blimps are a promising platform for aerial robotics and have been studied extensively for this purpose. Unlike other aerial vehicles, blimps are relatively safe and also possess the ability to loiter for long periods. These advantages, however, have been difficult to exploit because blimp dynamics are complex and inherently non-linear. The classical approach to system modeling represents the system as an ordinary differential equation (ODE) based on Newtonian principles. A more recent modeling approach is based on representing state transitions as a Gaussian process (GP). In this paper, we present a general technique for system identification that combines these two modeling approaches into a single formulation. This is done by training a Gaussian process on the residual between the non-linear model and ground truth training data. The result is a GP-enhanced model that provides an estimate of uncertainty in addition to giving better state predictions than either ODE or GP alone. We show how the GP-enhanced model can be used in conjunction with reinforcement learning to generate a blimp controller that is superior to those learned with ODE or GP models alone.

GP-UKF: Unscented kalman filters with Gaussian process prediction and observation models

This paper considers the use of non-parametric system models for sequential state estimation. In particular, motion and observation models are learned from training examples using Gaussian process (GP) regression. The state estimator is an unscented Kalman filter (UKF). The resulting GP-UKF algorithm has a number of advantages over standard (parametric) UKFs. These include the ability to estimate the state of arbitrary nonlinear systems, improved tracking quality compared to a parametric UKF, and graceful degradation with increased model uncertainty. These advantages stem from the fact that GPs consider both the noise in the system and the uncertainty in the model. If an approximate parametric model is available, it can be incorporated into the GP; resulting in further performance improvements. In experiments, we show how the GP-UKF algorithm can be applied to the problem of tracking an autonomous micro-blimp.