Tomasz Piotr Kucner

SPENCER: A Socially Aware Service Robot for Passenger Guidance and Help in Busy Airports

We present an ample description of a socially compliant mobile robotic platform, which is developed in the EU-funded project SPENCER. The purpose of this robot is to assist, inform and guide passengers in large and busy airports. One particular aim is to bring travellers of connecting flights conveniently and efficiently from their arrival gate to the passport control. The uniqueness of the project stems from the strong demand of service robots for this application with a large potential impact for the aviation industry on one side, and on the other side from the scientific advancements in social robotics, brought forward and achieved in SPENCER. The main contributions of SPENCER are novel methods to perceive, learn, and model human social behavior and to use this knowledge to plan appropriate actions in real-time for mobile platforms. In this paper, we describe how the project advances the fields of detection and tracking of individuals and groups, recognition of human social relations and activities, normative human behavior learning, socially-aware task and motion planning, learning socially annotated maps, and conducting empirical experiments to assess socio-psychological effects of normative robot behaviors.

Conditional transition maps: Learning motion patterns in dynamic environments

In this paper we introduce a method for learning motion patterns in dynamic environments. Representations of dynamic environments have recently received an increasing amount of attention in the research community. Understanding dynamic environments is seen as one of the key challenges in order to enable autonomous navigation in real-world scenarios. However, representing the temporal dimension is a challenge yet to be solved. In this paper we introduce a spatial representation, which encapsulates the statistical dynamic behavior observed in the environment. The proposed Conditional Transition Map (CTMap) is a grid-based representation that associates a probability distribution for an object exiting the cell, given its entry direction. The transition parameters are learned from a temporal signal of occupancy on cells by using a local-neighborhood cross-correlation method. In this paper, we introduce the CTMap, the learning approach and present a proof-of-concept method for estimating future paths of dynamic objects, called Conditional Probability Propagation Tree (CPPTree). The evaluation is done using a real-world dataset collected at a busy roundabout.

Towards occupational health improvement in foundries through dense dust and pollution monitoring using a complementary approach with mobile and stationary sensing nodes

In industrial environments, such as metallurgic facilities, human operators are exposed to harsh conditions where ambient air is often polluted with quartz, dust, lead debris and toxic fumes. Constant exposure to respirable particles can cause irreversible health damages and thus it is of high interest for occupational health experts to monitor the air quality on a regular basis. However, current monitoring procedures are carried out sparsely, with data collected in single day campaigns limited to few measurement locations. In this paper we explore the use and present first experimental results of a novel heterogeneous approach that uses a mobile robot and a network of low cost sensing nodes. The proposed system aims to address the spatial and temporal limitations of current monitoring techniques. The mobile robot, along with standard localization and mapping algorithms, allows to produce short term, spatially dense representations of the environment where dust, gas, ambient temperature and airflow information can be modelled. The sensing nodes on the other hand, can collect temporally dense (and usually spatially sparse) information during long periods of time, allowing in this way to register for example, daily variations in the pollution levels. Using data collected with the proposed system in an steel foundry, we show that a heterogeneous approach provides dense spatio-temporal information that can be used to improve the working conditions in industrial facilities.

Kinodynamic motion planning on Gaussian mixture fields

We present a mobile robot motion planning approach under kinodynamic constraints that exploits learned perception priors in the form of continuous Gaussian mixture fields. Our Gaussian mixture fields are statistical multi-modal motion models of discrete objects or continuous media in the environment that encode e.g. the dynamics of air or pedestrian flows. We approach this task using a recently proposed circular linear flow field map based on semi-wrapped GMMs whose mixture components guide sampling and rewiring in an RRT∗ algorithm using a steer function for non-holonomic mobile robots. In our experiments with three alternative baselines, we show that this combination allows the planner to very efficiently generate high-quality solutions in terms of path smoothness, path length as well as natural yet minimum control effort motions through multi-modal representations of Gaussian mixture fields.

Where am I? An NDT-based prior for MCL

One of the key requirements of autonomous mobile robots is a robust and accurate localisation system. Recent advances in the development of Monte Carlo Localisation (MCL) algorithms, especially the Normal Distribution Transform Monte Carlo Localisation (NDT-MCL), provides memory-efficient reliable localisation with industry-grade precision. We propose an approach for building an informed prior for NDT-MCL (in fact for any MCL algorithm) using an initial observation of the environment and its map. Leveraging on the NDT map representation, we build a set of poses using partial observations. After that we construct a Gaussian Mixture Model (GMM) over it. Next we obtain scores for each distribution in GMM. In this way we obtain in an efficient way a prior for NDT-MCL. Our approach provides a more focused then uniform initial distribution, concentrated in states where the robot is more likely to be, by building a Gaussian mixture model over potential poses. We present evaluations and quantitative results using real-world data from an indoor environment. Our experiments show that, compared to a uniform prior, the proposed method significantly increases the number of successful initialisations of NDT-MCL and reduces the time until convergence, at a negligible initial cost for computing the prior.

Enabling Flow Awareness for Mobile Robots in Partially Observable Environments

Understanding the environment is a key requirement for any autonomous robot operation. There is extensive research on mapping geometric structure and perceiving objects. However, the environment is also defined by the movement patterns in it. Information about human motion patterns can, e.g., lead to safer and socially more acceptable robot trajectories. Airflow pattern information allow to plan energy efficient paths for flying robots and improve gas distribution mapping. However, modeling the motion of objects (e.g., people) and flow of continuous media (e.g., air) is a challenging task. We present a probabilistic approach for general flow mapping, which can readily handle both of these examples. Moreover, we present and compare two data imputation methods allowing to build dense maps from sparsely distributed measurements. The methods are evaluated using two different data sets: one with pedestrian data and one with wind measurements. Our results show that it is possible to accurately represent multimodal, turbulent flow using a set of Gaussian Mixture Models, and also to reconstruct a dense representation based on sparsely distributed locations.

Probabilistic Air Flow Modelling Using Turbulent and Laminar Characteristics for Ground and Aerial Robots

For mobile robots that operate in complex, uncontrolled environments, estimating air flow models can be of great importance. Aerial robots use air flow models to plan optimal navigation paths and to avoid turbulence-ridden areas. Search and rescue platforms use air flow models to infer the location of gas leaks. Environmental monitoring robots enrich pollution distribution maps by integrating the information conveyed by an air flow model. In this paper, we present an air flow modelling algorithm that uses wind data collected at a sparse number of locations to estimate joint probability distributions over wind speed and direction at given query locations. The algorithm uses a novel extrapolation approach that models the air flow as a linear combination of laminar and turbulent components. We evaluated the prediction capabilities of our algorithm with data collected with an aerial robot during several exploration runs. The results show that our algorithm has a high degree of stability with respect to parameter selection while outperforming conventional extrapolation approaches. In addition, we applied our proposed approach in an industrial application, where the characterization of a ventilation system is supported by a ground mobile robot. We compared multiple air flow maps recorded over several months by estimating stability maps using the Kullback–Leibler divergence between the distributions. The results show that, despite local differences, similar air flow patterns prevail over time. Moreover, we corroborated the validity of our results with knowledge from human experts.

Learning to detect misaligned point clouds

Matching and merging overlapping point clouds is a common procedure in many applications, including mobile robotics, three-dimensional mapping, and object visualization. However, fully automatic po ...

A Dual PHD Filter for Effective Occupancy Filtering in a Highly Dynamic Environment

Environment monitoring remains a major challenge for mobile robots, especially in densely cluttered or highly populated dynamic environments, where uncertainties originated from environment and sensor significantly challenge the robot’s perception. This paper proposes an effective occupancy filtering method called the dual probability hypothesis density (DPHD) filter, which models uncertain phenomena, such as births, deaths, occlusions, false alarms, and miss detections, by using random finite sets. The key insight of our method lies in the connection of the idea of dynamic occupancy with the concepts of the phase space density in gas kinetic and the PHD in multiple target tracking. By modeling the environment as a mixture of static and dynamic parts, the DPHD filter separates the dynamic part from the static one with a unified filtering process, but has a higher computational efficiency than existing Bayesian Occupancy Filters (BOFs). Moreover, an adaptive newborn function and a detection model considering occlusions are proposed to improve the filtering efficiency further. Finally, a hybrid particle implementation of the DPHD filter is proposed, which uses a box particle filter with constant discrete states and an ordinary particle filter with a time-varying number of particles in a continuous state space to process the static part and the dynamic part, respectively. This filter has a linear complexity with respect to the number of grid cells occupied by dynamic obstacles. Real-world experiments on data collected by a lidar at a busy roundabout demonstrate that our approach can handle monitoring of a highly dynamic environment in real time.

Semi-supervised 3D place categorisation by descriptor clustering

Place categorisation; i.e., learning to group perception data into categories based on appearance; typically uses supervised learning and either visual or 2D range data. This paper shows place categorisation from 3D data without any training phase. We show that, by leveraging the NDT histogram descriptor to compactly encode 3D point cloud appearance, in combination with standard clustering techniques, it is possible to classify public indoor data sets with accuracy comparable to, and sometimes better than, previous supervised training methods. We also demonstrate the effectiveness of this approach to outdoor data, with an added benefit of being able to hierarchically categorise places into sub-categories based on a user-selected threshold. This technique relieves users of providing relevant training data, and only requires them to adjust the sensitivity to the number of place categories, and provide a semantic label to each category after the process is completed.