Xanthi S. Papageorgiou

Smart radiation sensor management

We developed two radiation mapping algorithms that can handle different situations based on prior information of the search area. The algorithms were developed in the framework of model-driven measurement, where a world model was used to drive measurement collection, and measurements were used to update the world model.We developed and experimentally tested a robotic implementation of two Bayesian-based radiation mapping strategies in two dimensions, using a commercially available desktop mobile robot fitted with a CsI radiation sensor. Our approach for implementing the Bayesian radiation mapping algorithms was to drive the robot over each segment of the search area, in real time, according to the radiation counts collected by the sensor. Future research directions include extensions to three-dimensional mapping; exploring and characterizing the tradeoffs between time efficiency, map confidence level, and utilization of prior knowledge information; as well as the implementation of Bayesian statistics for the online update of the world model.

Advances in Intelligent Mobility Assistance Robot Integrating Multimodal Sensory Processing

Mobility disabilities are prevalent in our ageing society and impede activities important for the independent living of elderly people and their quality of life. The goal of this work is to support human mobility and thus enforce fitness and vitality by developing intelligent robotic platforms designed to provide user-centred and natural support for ambulating in indoor environments. We envision the design of cognitive mobile robotic systems that can monitor and understand specific forms of human activity, in order to deduce what the human needs are, in terms of mobility. The goal is to provide user and context adaptive active support and ambulation assistance to elderly users, and generally to individuals with specific forms of moderate to mild walking impairment. To achieve such targets, a reliable multimodal action recognition system needs to be developed, that can monitor, analyse and predict the user actions with a high level of accuracy and detail. Different modalities need to be combined into an integrated action recognition system. This paper reports current advances regarding the development and implementation of the first walking assistance robot prototype, which consists of a sensorized and actuated rollator platform. The main thrust of our approach is based on the enhancement of computer vision techniques with modalities that are broadly used in robotics, such as range images and haptic data, as well as on the integration of machine learning and pattern recognition approaches regarding specific verbal and non-verbal gestural commands in the envisaged physical and non-physical human-robot interaction context.

Hidden Markov modeling of human normal gait using laser range finder for a mobility assistance robot.

For an effective intelligent active mobility assistance robot, the walking pattern of a patient or an elderly person has to be analyzed precisely. A well-known fact is that the walking patterns are gaits, that is, cyclic patterns with several consecutive phases. These cyclic motions can be modeled using the consecutive gait phases. In this paper, we present a completely non-invasive framework for analyzing a normal human walking gait pattern. Our framework utilizes a laser range finder sensor to collect the data, a combination of filters to preprocess these data, and an appropriately synthesized Hidden Markov Model (HMM) for state estimation, and recognition of the gait data. We demonstrate the applicability of this setup using real data, collected from an ensemble of different persons. The results presented in this paper demonstrate that the proposed human data analysis scheme has the potential to provide the necessary methodological (modeling, inference, and learning) framework for a cognitive behavior-based robot control system. More specifically, the proposed framework has the potential to be used for the recognition of abnormal gait patterns and the subsequent classification of specific walking pathologies, which is needed for the development of a context-aware robot mobility assistant.

Locally Computable Navigation Functions for Sphere Worlds

In this paper we present a new navigation function for a sphere world that can be computed locally with limited knowledge of the environment. By requiring smooth and not analytic NF, the effect of each obstacle is exactly nullified outside a sensing zone around the obstacle (the only required parameter is the width of the sensing zone). This allows the computation of the navigation function using information from a single obstacle each time. We present simulations to verify the validity of this approach.

Hidden markov modeling of human pathological gait using laser range finder for an assisted living intelligent robotic walker

The precise analysis of a patients or an elderly persons walking pattern is very important for an effective intelligent active mobility assistance robot. This walking pattern can be described by a cyclic motion, which can be modeled using the consecutive gait phases. In this paper, we present a completely non-invasive framework for analyzing and recognizing a pathological human walking gait pattern. Our framework utilizes a laser range finder sensor to detect and track the human legs, and an appropriately synthesized Hidden Markov Model (HMM) for state estimation, and recognition of the gait patterns. We demonstrate the applicability of this setup using real data, collected from an ensemble of different elderly persons with a number of pathologies. The results presented in this paper demonstrate that the proposed human data analysis scheme has the potential to provide the necessary methodological (modeling, inference, and learning) framework for a cognitive behavior-based robot control system. More specifically, the proposed framework has the potential to be used for the classification of specific walking pathologies, which is needed for the development of a context-aware robot mobility assistant.

Motion tasks for robot manipulators on embedded 2-D manifolds

In this paper we present a methodology to drive the end effector of a robotic manipulator across the surface of an object in the workspace. Three typical tasks are considered, namely stabilization of the end effector over the objects surface, motion planning and eventually trajectory tracking of the end effector across the objects surface. The proposed controllers utilize navigation functions and are based on the belt zone vector fields concept. The derived dynamic controllers are realized using an integrator backstepping methodology. The derived feedback based controllers guarantee global convergence and collision avoidance. The closed form solution provides fast feedback rendering the methodology particularly suitable for implementation on real time systems. The properties of the proposed methodology are verified through non-trivial computer simulations

Motion Planning and Trajectory Tracking on 2-D Manifolds embedded in 3-D Workspaces

In this paper we present a methodology that drives and stabilizes a robotic agent moving in a three dimensional environment, to a 2-dimensional manifold embedded in the workspace. Once the agent reaches the manifold, depending on the application, it performs a motion planning or a trajectory tracking task. Appropriately constructed belt-zone vector fields guarantee that the agent will not depart the 2-D manifold proximity area, while carrying out the motion planning or trajectory tracking task. The derived closed form feedback control law guarantees global convergence and collision avoidance. The properties of the proposed algorithm are verified through non-trivial computer simulations.

Experimental validation of human pathological gait analysis for an assisted living intelligent robotic walker

A robust and effective gait analysis functionality is an essential characteristic for an assistance mobility robot dealing with elderly persons. The aforementioned functionality is crucial for dealing with mobility disabilities which are widespread in these parts of the population. In this work we present experimental validation of our in house developed system. We are using real data, collected from an ensemble of different elderly persons with a number of pathologies, and we present a validation study by using a GaitRite System. Our system, following the standard literature conventions, characterizes the human motion with a set of parameters which subsequently can be used to assess and distinguish between possible motion disabilities, using a laser range finder as its main sensor. The initial results, presented in this work, demonstrate the applicability of our framework in real test cases. Regarding such frameworks, a crucial technical question is the necessary complexity of the overall tracking system. To answer this question, we compare two approaches with different complexity levels. The first is a static rule based system acting on filtered laser data, while the second system utilizes a Hidden Markov Model for gait cycle estimation, and extraction of the gait parameters. The results demonstrate that the added complexity of the HMM system is necessary for improving the accuracy and efficacy of the system.

Towards locally computable polynomial navigation functions for convex obstacle workspaces

In this paper we present a polynomial navigation function (NF) for a sphere world that can be constructed almost locally, with partial knowledge of the environment. The presented navigation function is C2 and as a result the computational complexity is very low, while the construction uses local knowledge and information. Moreover, an almost locally computable diffeomorphism between convex obstacles and spheres is presented, allowing the NF scheme to be used in a workspace populated by convex obstacles. Our approach is not strictly local in the epsiv sense, i.e., the field around a point is not influenced only by an e region around the point, but rather it is local in the sense that the NF around each obstacle is influenced only by the obstacle and the adjacent obstacles. In particular, we require, in the vicinity of an obstacle, the distance between the obstacle and the adjacent obstacles. Simulations are presented to verify this approach.

Motion tasks for robot manipulators subject to joint velocity constraints

We present a methodology to steer the end effector of a robotic manipulator, which is constrained in terms of joint rates, on the surface within the workspace. We develop controllers for stabilizing the end effector to a point, and for tracking a trajectory on this surface, while respecting the input constraints. We show that the resulting closed loop system is uniformly asymptotically stable and we verify our analytical development with computer simulations.