Localization of Autonomous Telepresence Robots

Duration: 01.09.2014 - 30.09.2016
Project lead: Prof. Dr.-Ing Horst Hellbrück
Staff: Grigori Goronzy, Mathias Pelka


For the currently emerging services and applications in the field of telepresence systems, a single-axis, self-balancing robot system is developed. For this purpose, the competence center CoSA of the University of Applied Sciences Lübeck develops a cost effective solution for a localization system which is evaluated with a reference system for the use case of a telepresence robot.

nolo design GmbH is responsible for the robot platform and the remote control functions. nolo design has many years of experience in the development of self-balancing personal transporters, which only require a small footprint and thus can have a very compact design.


The new telepresence robot can autonomously navigate and it can be remote controlled as well. Existing localization systems that are needed for autonomous navigation exceed the cost of a telepresence robot many times. Therefore, a cost-effective solution is the key to the success of the project. CoSA is responsible for modeling the robot odometry, design and implementation of the localization and the reference point systems which are necessary to determine the correct position of the robot. A high accuracy up to a few centimeters is sought.


As first part of the project, odometry with self-balanced robot platforms will be evaluated and implemented. In addition, different approaches for absolute localization by means of anchor points are evaluated. Absolute localization and odometry are finally combined to obtain a suiable localization result for an indoor navigation solution.

Project Partners

nolo design GmbH, Fahrenkrug

Förderkennzeichen: KF3177202PR4




Refereed Articles and Book Chapters
[2016] Survey of challenges and towards a unified architecture for location systems (Mathias Pelka, Horst Hellbrück), In Journal of Network and Computer Applications, volume 67, 2016. [bib] [pdf] [abstract]
Abstract Localization is a key aspect of emergent applications in the medical, industrial and consumer field. In this article we survey state of the art, identify current challenges and issues for localization systems and suggest a unified layered architecture. The analysis reveals that challenges cannot be addressed in an isolated manner for example, energy consumption is tied to the choice of algorithm and employed hardware. To separate various challenges and investigate them independently, we propose the concept of position providers. Position providers in the lower layers allow abstraction of positioning methods, positioning algorithms and positioning hardware. Thereby, a position provider encapsulates methods, algorithms and hardware. Furthermore, we suggest a classification of position providers inspired by related work. We propose a unified architecture for location systems which uses positioning and integration layers as main building blocks.
[2016] Iterative approach for anchor configuration of positioning systems (Mathias Pelka, Grigori Goronzy, Horst Hellbrück), In ICT Express, volume 2, 2016. [bib] [pdf] [abstract]
With anchor positions and measurements of distances between an object and anchors, positioning algorithms calculate the position of an object, e.g. via lateration. Positioning systems require calibration and configuration prior to operation. In the past, approaches employed reference nodes with GPS or other reference location systems to determine anchor positions. In this article, we propose an approach to determine anchor positions without prior knowledge. We evaluate our approach with simulations and real data based on the Decawave DW1000 radio and show that the error is proportional to the mean error of the distance estimation.
Refereed Conference Papers
[2016] S-TDoA - Sequential Time Difference of Arrival - A Scalable and Synchronization Free Approach for Positioning (Mathias Pelka, Horst Hellbrück), In IEEE Wireless Communications and Networking Conference, 2016. [bib] [abstract]
In the past various solutions for localization evolved to productive usage for wireless applications. These solutions are robust, precise and energy efficient. However, scalability, complexity and flexibility are still open issues. Especially, supported number of objects or update rates for localization are still limiting factors for the usage of the systems. In this work we suggest an approach called S-TDoA which stands for sequential Time Difference of Arrival that supports unlimited number of objects and high update rates. The key concept is a sequential triggering of anchors that send periodic messages. Tags determine their position by listening to the anchor messages and measuring time intervals. Additionally, this approach enhances security because tags are not visible as they do not send messages. We implement and evaluate S-TDoA in a localization system based on UWB-RF- Chips. The preliminary results demonstrate the advantages of our implementation regarding scalability and update rates as well as privacy.
[2016] Impact of Altitude Difference for Local Positioning Systems and Compensation with Two-Stage Filters (Mathias Pelka, Grigori Goronzy, Horst Hellbrück), In 2016 International Conference on Localization and GNSS, 2016. [bib] [abstract]
In range-based positioning systems, an altitude difference between tag and reference plane causes errors in two- and three-dimensional positioning. We analyze how these errors reduce accuracy of Local Positioning Systems (LPS) and show how compensation of the altitude difference improves performance of positioning. In this paper, we consider the availability of additional altitude information and transform the three-dimensional positioning problem into a two-dimensional problem. We provide algorithms for time-based positioning systems with a two-stage estimator for Two-Way Ranging and Time Difference of Arrival and incorporate additional altitude information. We simulate our approach for altitude difference compensation and provide an evaluation based on a Ultra-Wideband (UWB) radio with ranging capability and a barometric sensor for additional altitude information. A comparison is then made between our approach and standard solutions such as the Extended Kalman filter and the Unscented Kalman filter. Finally, the successful decrease in the positioning error for two- and three-dimensional positioning system, using the system disclosed herein, is illustrated. Based on our analysis, we derive practical solutions to deal with altitude differences for positioning systems.
[2016] Introduction, Discussion and Evaluation of Recursive Bayesian Filters for Linear and Nonlinear Filtering Problems in Indoor Localization (Mathias Pelka, Horst Hellbrück), In The Seventh International Conference on Indoor Positioning and Indoor Navigation, 2016. [bib] [abstract]
Linear and nonlinear filtering for state estimation (e.g. position estimation or sensor fusion) for indoor positioning and navigation applications is a challenging task. Sensor fusion becomes more important with cost-effective sensors being readily available. However, state estimation with recursive Bayesian filters for sensor fusion and filtering are difficult to apply. We present an overview for the general Bayesian filter and derive the most commonly used recursive Bayesian filters, namely the Kalman, extended Kalman and the unscented Kalman filter along with the particle filter. The later Kalman filters are extension of the original Kalman filter, which are able to solve nonlinear filtering problems. The particle filter is also able to solve nonlinear filtering problems. We evaluate the recursive Bayesian filters for linear and nonlinear filtering problems for sensor fusion from relative dead reckoning positioning data and absolute positioning data from an UWB positioning system. We discuss and evaluate performance and computational complexity and provide recommendations for the use case of the recursive Bayesian filters.
[2016] QRPos: Indoor Positioning System for Self-Balancing Robots based on QR Codes (Grigori Goronzy, Mathias Pelka, Horst Hellbrück), In The Seventh International Conference on Indoor Positioning and Indoor Navigation, 2016. [bib]
[2016] Investigation of Anomaly-based Passive Localization with Received Signal Strength for IEEE 802.15.4 (Marco Cimdins, Mathias Pelka, Horst Hellbrück), In The Seventh International Conference on Indoor Positioning and Indoor Navigation, 2016. [bib] [abstract]
Localization has important applications, for instance intrusion detection and elderly care. Such applications benefit from Device-free passive (DfP) localization systems, which employ received signal strength measurements (RSSM) to detect and track entities that neither participate actively in the localization process nor emit signals actively. RSSMs include received signal strength indicator (RSSI), energy detection (ED) and link quality indicator (LQI) measurements. This paper compares different packet-based RSSMs for DfP localization and presents detection results of a DfP anomaly-based detection system employed by IEEE 802.15.4 compliant devices. Furthermore, we investigate techniques for anomaly detection with continuous RSSI measurements.
Refereed Workshop Papers
[2017] Weighted Online Calibration for Odometry of Mobile Robots (Grigori Goronzy, Horst Hellbrück), In IEEE ICC Workshop on Advances in Network Localization and Navigation (ANLN), 2017. [bib]
[2016] Comparison of wired and wireless synchronization with clock drift compensation suited for U-TDoA localization (Swen Leugner, Mathias Pelka, Horst Hellbrück), In 13th Workshop on Positioning, Navigation and Communication, 2016. [bib] [abstract]
Indoor localization with Uplink Time Difference of Arrival (U-TDoA) provides good scalability, high updates rates and high accuracy. However, clock errors lead to localization errors and synchronization is important. In this paper, we design and implement wired and wireless synchronization and provide a comparison between them. We design and implement a wireless synchronization with clock drift compensation. For wired and wireless synchronization, we discuss reasons for clock deviation that lead to localization errors. We evaluate both approaches in a U-TDoA measurement setup. Finally, we provide recommendations for wired and wireless synchronization.
[2015] Mobile Robot Seamless Localization with Localization Optimized QR Codes (Mathias Pelka, Daniel Neckel, Horst Hellbrück), In 12th Workshop on Positioning, Navigation and Communication, 2015. [bib] [abstract]
Indoor navigation is a prerequisite for new emerging applications for autonomous mobile robots. Additionally to the location of a robot, the orientation is important for these applications. Furthermore, a solution to this localization problem should be inexpensive and easy extensible for new areas of a building. We propose inexpensive optical landmarks based on localization optimized Quick Response (QR) code for localization of the landmark within an image to reduce computational cost. We further specify the error correction level, border, and size of the QR code for optimal localization. The proposed QR code combines GPS coordinates and local coordinates which allows seamless integration of our approach. We perform image processing to estimate the distance and orientation of a mobile robot with respect to the localization optimized QR code. To evaluate our approach we implemented the approach in an Android application and measured the performance in experiments. Additionally, we suggest a method to retrieve more accurate GPS information based on the measured orientation and distances. Our implementation achieves update rates of up to 3 Hz and an accuracy of 1 cm
Other Publications
[2016] Investigation of Anomaly-based Passive Localization with IEEE 802.15.4 (Marco Cimdins, Mathias Pelka, Horst Hellbrück), Technical report, RWTH Aachen University, 2016. [bib] [pdf]
[2015] Position Calculation with Least Squares based on Distance Measurements (Mathias Pelka), Technical report, Fachhochschule Lübeck, 2015. [bib] [pdf] [abstract]
Position estimation based on distances is a well understood problem. This document describes a simple way to linearize the position equation. Based on the linearization the problem is solved step by step using least squares. This paper includes an example implementation in Matlab.
Powered by bibtexbrowser