` Intelligent and Mobile Robotics

Current Projects

SafeLog

MUAVET

Kassandra

CAK3

SR4RL


SafeLog - Safe human-robot interaction in logistic applications for highly flexible warehouses

The European market for e-commerce is growing rapidly, with more than 16% just in the year 2014. With the internationalization of distribution chains, the key for success lies within efficient logistics. In such facilities, goods for the end-user or products in the B2B sector are stored, commissioned and shipped. To manage the supply chains, many new warehouses have been erected and more will follow. With the growing markets, the need for larger warehouses and their automation increases. To advance the position of the European trade sector, technical restrictions on the size of warehouses should be avoided and new automation paradigm should be implemented to ensure their efficient operation. Therefore the European robotic and automation companies should be able to provide appropriate solutions, making scalable systems and scalable software mandatory. Current automation solutions based on strict separation of humans and robots cannot provide such efficient operation of large warehouses. SafeLog aims to overcome this issue by enabling much more efficient warehouse concepts joining human and robot workforce. Given that, the overall objective of SafeLog is the conception and implementation of a large-scale flexible warehouse system which enables safe and efficient collaboration of humans and robots in the same area and at the same time. On the way to reach this overall objective SafeLog will develop, integrate and test: (1) a holistic and certifiable safety concept based on the safety vest, which allows the collaboration of robots and humans in a flexible warehouse system, (2) planning and scheduling algorithms for a heterogeneous fleet manager, which allow the adhoc reactive planning and scheduling for human and robot workforce in a flexible warehouse system, and (3) augmented reality based interaction strategies to support workers in a robotized warehouse system with information about their current task and environment.

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 688117.



MUAVET - Multi-Aerial Vehicles Experimentation Testbed

The project entitles Multi-Aerial Vehicles Experimentation Testbed (MUAVET) and targets design of and development of an experimental test-bed for multi-robot experimentation. The system in its’ early outfit targets a flock of autonomous micro-aerial vehicles (MAVs) for experimentation in the search and inspection task domains, with a focus onto future extensions to heterogenous types of robots (ground and aerial vehicles, providing diverse resources) for research of advanced communication technologies, resource sharing, and to study cooperative and collective robotic problems. The prospective application are heading environmental inspection and control and S&R scenarios.

A joint project of CTU/CIIRC and King Abdulaziz University, SA


KASSANDRA – Multi-camera vehicles' undercarriage security scanner

The project focuses on the development of easily installable security car scanner, which provides accurate 3D reconstruction of the chassis with a color image and on the research of methods for evaluating this data using methods of artificial intelligence, image processing and pattern recognition. The system assigns the data obtained from the scanner to the car's registration number and compares it with stored reference data and highlights found differences so that the operator can evaluate risk.

A joint project of VOP CZ and CTU/CIIRC funded by Ministry of Interior of the Czech Republic under the project number VI2VS/461.


CAK3 - Centre for Applied Cybernetics 3

Centre for Applied Cybernetics 3 is a project of the Competence Centres Programme supported by the Technology Agency of the Czech Republic. The project is a natural continuation of Centre for Applied Cybernetics 1, which was supported within the Programme of Research Centres of the Ministry of Education, Youth and Sport, Czech Republic during 2000 – 2004 (LN00B096) and of Centre for Applied Cybernetics 2, which was supported within the Programme of Applied Research Centres of the Ministry of Education, Youth and Sport, Czech Republic during 2005 – 2011 (1M0567). The Centre makes use of the existing infrastructure and the experience accumulated throughout the years. However, its strategic research agenda is more focused and shifted towards the needs of the industry. The Centre involves the foremost research teams nationwide from public universities and high-tech companies.

IMR is responsible for two workpackages: WP6 - components of a robotic system for human-robot interaction (solved in 2012-2015) and WP17- Tools for human robot interaction (to be solved in 2016-2019. WP6 was focused on research and development of intelligent systems for HRI (human-robot interaction), specifically robotic systems for autonomous inspection, exploration, and surveillance, wich incorporates robust navigation in real environments. Within this task, a tool for autonomous exploration in a polygonal domain EAPD has been created. WP17 aims to design and realize appropriate approaches to autonomous failure detection in robotic systems and non-assistive recovery from them.


SR4RL - Symbolic Regression for Reinforcement Learning in Continuous Spaces

Reinforcement Learning (RL) algorithms can optimally solve dynamic decision and control problems in engineering, economics, medicine, artificial intelligence and other disciplines. However, state-of-the-art RL methods still have not solved the transition from a small set of discrete states to large or continuous state spaces. They have to rely on function approximators, such as radial basis functions, to represent the value function and policy mappings. The choice of a suitable approximator, in terms of its structure is a difficult step which always requires significant trial-and-error tuning. The goal of the project is to automate the search process for the value function and policy approximator, so that it becomes an integral part of the learning procedure. We will further develop symbolic regression, a technique based on genetic programming, so that it can automatically find analytic functions representing the policy and value function mappings with the RL framework. The result will be a new class of RL methods suitable for continuous, high-dimensional state and action spaces.

The IMR Group is responsible of the experimental validation of the developed algorithms. The first experimental setup consists of autonomous modules that can be connected to form bodies of various shapes. Modular robots can be used in applications like space exploration, search and rescue missions, and in object manipulation. The main advantage of modular robots lies in the possibility of forming versa- tile morphologies that can move in diverse terrains. The second experimental setup is based on SyRoTek – an e-learning platform for mobile robotics, artificial intelligence and control, developed at CTU Prague. SyRoTek provides remote access to a set of fully autonomous mobile robots equipped with standard robotic sensors (laser and sonar range finders, cameras, odometry, etc.) and placed in a restricted area with dynamically reconfigurable obstacles.

This project has received funding from the Grant Agency of Czech Republic under grant agreement No GA15-22731S.


SyRoTek - System for robotic e-learning (CEMRA)

The project aims to extend the existing robotic e-learning platform SyRoTek developed at Czech Technical University in Prague and providing distant access to real mobile robots by creation of a teaching material. This material includes an on-line introductory course consisting of a set of fundamental robotic tasks, interactive introduction to the SyRoTek system, and interactive demo applications introducing key mobile robotics concepts. All the material will be in the form of web pages, which will contain theoretical description of each task/problem, interface supporting user's own work with real robots, and a quiz verifying user's understanding of the problem. A tool for automatic evaluation of behaviour of user's code on real robots will be also developed and integrated into the system. The main goal of the project is to enlarge the SyRoTek community by allowing its users to learn the system independently and then work with it without need of interaction with a teacher or an administrator. We believe that this will attract not only universities and other educational institutions to use the system for their education but also researchers from robotics and related areas to perform experiments with a real robotic hardware.

This project has received funding from the IEEE Robotics and Automation Society under the program CEMRA (Creation of Educational Material in Robotics and Automation).


Past Projects

SyRoTek

The aim of the SyRoTek (“System for robotic e-learning”) project is to research, design, and develop novel methods and approaches to building a multi-robot system for distance learning. The system allows its remote users to get acquainted with algorithms from areas of modern mobile and collective robotics, artificial intelligence, control, and many other related domains. Advanced users can develop their own algorithms and monitor the behaviour of these algorithms on-line during real experiments. The proposed system reduces the development process and allows a wide spectrum of both individuals and institutions to work with real robotic equipment.

The main components of the robotic platform – mobile robots – are moving inside a restricted area, which also contains other elements like obstacles or objects related to objectives of the actually solved task. Moreover, several sensors (infrared, sonars, cameras, etc.) are used to gather information about the actual status of the playfield and particular objects on it. Some sensors are placed onboard the robots, while others are stand-alone, getting a global overview of the playfield status. The user can not only observe the gathered data using the internet interface, but also control the robots in real-time. Unlike existing e-learning robotic systems developed in the world in which the user can only tele-operate robots, behaviour of the robots in the SyRoTek system can be modified, while the system allows to run algorithms developed by the user.


Symbrion/Replicator

The main focus of these projects is to investigate and develop novel principles of adaptation and evolution for symbiotic multi-robot organisms based on bio-inspired approaches and modern computing paradigms. Such robot organisms consist of super-large-scale swarms of robots, which can dock with each other and symbiotically share energy and computational resources within a single artificial-life-form. When it is advantageous to do so, these swarm robots can dynamically aggregate into one or many symbiotic organisms and collectively interact with the physical world via a variety of sensors and actuators. The bio-inspired evolutionary paradigms combined with robot embodiment and swarm-emergent phenomena, enable the organisms to autonomously manage their own hardware and software organization. In this way, artificial robotic organisms become self-configuring, self-healing, self-optimizing and self-protecting from both hardware and software perspectives. This leads not only to extremely adaptive, evolve-able and scalable robotic systems, but also enables robot organisms to reprogram themselves without human supervision and for new, previously unforeseen, functionality to emerge. In addition, different symbiotic organisms may co-evolve and cooperate with each other and with their environment.

SYMBRION, SYMBRION Enlarged EU

Duration: 01.02.2008-31.01.2013

funded by FP7: FET Proactive Intiative: PERVASIVE ADAPTATION

  • platform for exploring artificial evolution and pervasive evolve-ability
  • extremely powerful computational on-board resources
  • support for artificial immunology and embryology
  • large number of light modules
  • powerful on-boad and off-board simulator

REPLICATOR

Duration: 01.03.2008-28.02.2013

funded by FP7: Cognitive Systems, Interaction and Robotics

  • intelligent, reconfigurable and adaptable “carrier” of sensors (sensors network)
  • sensors- and communication-rich platform
  • high-reliable in open-end environment
  • medium number of heavy modules

COLOS - Control and Localization for Swarms of Low-cost Autonomous Robots

The project COLOS aims to integrate principles and theoretical background of swarm behaviour with methodology, theory describing cooperative localization of autonomous robots and principles of self-organizing adaptation in a new unique way leading to a new flexible stand-alone system. It will enable applicability of SWARM robotics in realistic scenarios of surveillance and reconnaissance. Here, we can find an inspiration in nature: individuals in big school of fish, clusters of insect or flocks of birds cannot directly communicate with neighbors due to surrounding noise and their relative localization is realized through observation of neighbors. Our idea is to transfer similar concepts to localization in robotics, mainly in applications of teams of autonomous low-cost helicopters (small quad-rotor helicopters equipped with simple panoramatic cameras).

The project consortium consists of two partners. The Czech (IMR - Intelligent Mobile Robotics group at the Gerstner Laboratory, CTU in Prague) partner is continuously developing a framework that is focused on main principles how to increase the localization precision leading to reliable navigational techniques. Moreover, it is the leading body of SOM application to the multi-goal path planning problems. Within the project, their approach and developed theory are used as a core of arising system for localization of swarms and their shape forming. The US partner (GRASP laboratory of Department of Mechanical Engineering and Applied Mechanics, University of Pennsylvania) is one of leading bodies in area of control of cooperating mobile robots and quad-rotor helicopters. Within the project, their experience with control of cooperating UAVs and knowledge of stability analysis is utilized to develop robust system applicable in scenarios of swarm robotics that cannot be realized with existing methods.


GAČR - Stabilization of μ-UAV swarms under decentralized relative localization.

In this project, a methodology enabling outdoor deployment of swarms of μ-Unmanned Aerial Vehicles (μ-UAVs) will be established and experimentally verified. Principles of a decentralized relative localization of neighboring particles will be designed and integrated to swarm behaviors with an aim to keep reciprocal visibility between neighbors. This enables to employ μ-UAV swarms outside laboratories equipped by a precise positioning system. Besides, a concept of adaptively evolving swarm behaviors will be established to decrease relative localization uncertainty. To enable multi-robot applications, theoretical principles of determining desired shapes of μ-UAV swarms will be designed. Finally, a decentralized collective decision making mechanism will be established with a theory identifying necessary assumptions of the switching between different swarm behaviors. This research will be aimed at a study of observed autonomous behaviors of μ-UAV swarms. All methods will be accompanied with proofs of stability or convergence, and verified via numerical simulations and outdoor experiments.


  • SGS - PhD and Master student projects
  • ...

MŠMT Kontakt Cognitive learning autonomous robots for services (joint project with Universidad de Buenos Aires)

The goal of the projectis to increase scientific and proffesional competence and to acquire new knowledge in the field of the autonomous mobile robotic systems and to develop an autonomous navigation system, which will allow robots to carry out useful taks (e.g. pick and delivery, surveillance, inspection) in everyday environments.


PeLoTe

The project was targeted on concepts of creating the presence feature via remote control and navigation, as well as knowledge sharing in combined communities of living and nonliving entities. The foreseen applications include next generation teleoperation and telediagnosis systems incorporating both humans and (semi)autonomous robots, design of highly realistic manmachine interfaces, allowing to accomplish complex teleoperation tasks.

The aim of the PELOTE project was to investigate the following key items:

  • Integration of humans and robots: methods, algorithms and methodologies for robust, efficient and flexible integration of autonomous robots and humans.
  • Humans as telematic entities: personal assistance and navigation systems (PENA) supporting humans to become the remote end entities in a teleoperated task. This concept extends their capabilities by navigation, assistance and sharing of data from remote knowledge bases.
  • Robots as telematic entities: research foundations of entities’ ability to maintain knowledge from another entity and/or from the teleoperation centre in order to provide a robust, flexible and autonomous level of cooperation.
  • Cooperation and coordination: comprises study and experimentation with efficient mechanisms and robust concepts for cooperation and collaboration during the exploration of an unknown environment. Development of cooperation strategies between multiple autonomous robots and human teleoperators, in order to take advantage of synergies in generating a joint knowledge base.
  • Communication: desires investigation of mechanisms for robust data transfer in telepresence and telediagnosis tasks.

MorBot

The goal of this project is a creation of a robotic platform called morbot dedicated for support of mobile robotics education. It enables basic introduction to typical problems relating with this theme, like robot localisation or environment mapping. There are some special demands for the robot like bump resistance, adjustable skeleton and sensors mounting or possibility of platform integration to the Player system. The goal of this project is also a creation of supporting software for teaching of mobile robotics. A part of support is configuration of the main computer of the robotic platform, which is not only settings of peripheral devices but mainly setup of cross-compilation environment for compiling the Player framework with implemented Morbot accessing module. An advantage of using the Player environment is simple application migration from a robot simulator to the real robot platform Morbot.


VenZeTmy

The aim of the project was to develop a new kind of an navigation aid for visually impaired. The project was managed by Integrace. We worked on the obstacle detection system, which was based on a stereo camera.

The aid consists of the sensor for obstacle detection and a tactile display which is integrated to a shirt. The tactile display has 7x3 vibrating actuators. The position of an obstacle in front of the user determines which actuator is activated. This allows the user to estimate positions of the surrounding obstacles.


CAK and CAK II

Information technologies are a driver of technology progress and society development. Future developments will most likely occur in the current bottleneck of information technologies, which is the production, representation, processing, interpretation, and application of information about the real world. Cybernetics integrates the fields that continuously widen this bottleneck. Cybernetics provides mathematical models of complex phenomena and applies these models in cognition, control, and decision-making. That is why cybernetics has been selected among the priorities of the National Research Programme.Cybernetics is a multidisciplinary field. Consequently, the individual disciplines have sometimes been progressing in isolation. Similarly, the results obtained have been applied in different industries, sometimes with a certain waste of effort.The Centre for Applied Cybernetics integrates the national research potential in this field. The proposed project links well-established research teams in academia with agile, small and medium industrial enterprises. The research done in the Centrel includes a spectrum of activities, from basic and applied research to industrial realizations. The companies involved are able to use the research results before they are published in journals and discussed at conferences. At the same time, the industrial companies involved will be able to influence the research focus at the Centre.


MSM6840770038 – Research plan (CZ: Výzkumný záměr) Decision-making and management for industrial production I-III

The research project focuses on the latest trends in the field of automatic control and modeling, innovative methods of artificial intelligence, machine perception and learning. The objectives of the project include basic research and applied research in order to gain new knowledge for future use in industrial practice. In particular the subject of research is covered by the automatic control in real time at the lowest level over the algorithms for optimal and robust control in high level, then through planning and simulation, systems integration, database systems to business competitiveness issues. Project aims to incorporate issues of intelligent systems for industrial production based on computer vision, robotics and new technical approaches, diagnosis and systemic safety issues associated with the transfer of industrial data. The role of mathematical support for the formalization and algorithmic issues of interoperability is further highlighted. The main objective is to achieve competitive results on a world-wide scale.


GAČR - Simulator of train position locator

The objective of the project was to design and verify the simulator of the train position locator based on satellite navigation (GNSS) and inertial navigation (INS) . The simulator is able to simulate functions of a real locomotive locator using input data recorded from sensors of the train position locator during test ride on real track . The functions to be verified are the following: 1) accurate position determination during train rooting on switches, 2) accurate position determination on track segments with limited GNSS satellite "visibility" (tunnels, track cuttings, …), 3) simulation of the train positioning under different speed profiles. The main benefit of the project consists in a fact, that the simulator helps to optimize the locator architecture and its parameters with respect to railway safety critical applications – interlocking and train control systems. The simulator also enables to simulate hazardous states what is impossible under normal operational conditions. The application of the simulator reduces a cost and time necessary for the locator development and trials.


  • SGS - PhD and Master student projects
  • IGS - PhD student projects
  • Kontakt - joint project with University of Ljubljana
  • ...