Frank Hoeller

A Component-Based Approach to Visual Person Tracking from a Mobile Platform

In this article, we present a component-based visual tracker for mobile platforms with an application to person tracking. The core of the technique is a component-based descriptor that captures the structure and appearance of a target in a flexible way. This descriptor can be learned quickly from a single training image and is easily adaptable to different objects. It is especially well suited to represent humans since they usually do not have a uniform appearance but, due to clothing, consist of different parts with different appearance. We show how this component-based descriptor can be integrated into a visual tracker based on the well known Condensation algorithm. Several person tracking experiments carried out with a mobile robot in different laboratory environments show that the system is able to follow people autonomously and to distinguish individuals. We furthermore illustrate the advantage of our approach compared to other tracking methods.

RoSe - A framework for multicast communication via unreliable networks in multi-robot systems

Multi-robot systems obtain their performance advantages from close collaboration of the participating robot systems. As the demand for such cooperation increases, the aspect of wireless communication is getting more and more important. To cope with the challenges of the wireless communication we propose an easy to use communication framework for a multi-robot system. This framework is designed for wireless networks, therefore it expects unreliable communication and can even cope with complete network separations. Using multicast as the default communication scheme, the framework can directly benefit from the broadcast nature of the wireless medium if such an optimisation is supported by the lower network protocols. We therefore introduce our multicast capable wireless routing protocol, which supports this optimisation. The software is implemented and frequently used in our experimental multi-robot system.

The ROS Multimaster Extension for Simplified Deployment of Multi-Robot Systems

This tutorial chapter describes how to set up a multi-robot system in ROS with the multimaster_fkie package. The package adds ROS support for multiple hosts, which can be added and removed from the network at any time without affecting the remaining nodes. The presented multi-master extension works with the unmodified ROS master and does not change the way ROS nodes communicate or establish connections with each other. Thus, the multi-robot system remains fully compatible with a single-master ROS system. It is easy to set up and execute the ROS masters independently on each robot. The multi-master extension takes care of synchronization and merges the masters into a unified network view. For better usability, the package includes a graphical user interface for monitoring, configuration and control of the ROS components. The latest version can be downloaded from https://github.com/fkie/multimaster_fkie. You can also install the package from http://packages.ros.org. The multimaster_fkie package works with all ROS versions since groovy.

Benchmark of 6D SLAM (6D Simultaneous Localisation and Mapping) Algorithms with Robotic Mobile Mapping Systems

Abstract This work concerns the study of 6DSLAM algorithms with an application of robotic mobile mapping systems. The architecture of the 6DSLAM algorithm is designed for evaluation of different data registration strategies. The algorithm is composed of the iterative registration component, thus ICP (Iterative Closest Point), ICP (point to projection), ICP with semantic discrimination of points, LS3D (Least Square Surface Matching), NDT (Normal Distribution Transform) can be chosen. Loop closing is based on LUM and LS3D. The main research goal was to investigate the semantic discrimination of measured points that improve the accuracy of final map especially in demanding scenarios such as multi-level maps (e.g., climbing stairs). The parallel programming based nearest neighborhood search implementation such as point to point, point to projection, semantic discrimination of points is used. The 6DSLAM framework is based on modified 3DTK and PCL open source libraries and parallel programming techniques using NVIDIA CUDA. The paper shows experiments that are demonstrating advantages of proposed approach in relation to practical applications. The major added value of presented research is the qualitative and quantitative evaluation based on realistic scenarios including ground truth data obtained by geodetic survey. The research novelty looking from mobile robotics is the evaluation of LS3D algorithm well known in geodesy.

CBRNE hazard detection with an Unmanned vehicle

This paper presents a research project for a CBRNE hazard detection robot prototype for the German military forces. One of the main design principles is the usage of commercially available sensors that can be exchanged and upgraded easily without touching the underlying robot platform.

Autonomous reconnaissance and surveillance in urban structures - Eurathlon 2013

In this paper we propose an integrated hard- and software system for autonomous exploration and mapping of dilapidated buildings. The system is based on well understood approaches towards SLAM and exploration. Because of the real world nature of the application, a fragile wireless connection and difficult obstacles like staircases were taken into account during system design. Additionally, a drop-off system for Wi-Fi relays was introduced in order to increase the communication range. Further, a homing function is added to safely explore beyond radio coverage. Negative obstacles are handled using a tilted laserscanner. A semi-automatic stair climbing mechanism allows to travel between floors. The system is interconnected with a control station, giving an operator a fine grained control over the mission without disturbing the exploration and mapping task. The system was successfully tested at the Eurathlon 2013 robotics competition under real world conditions.