Jaroslav Rozman

Grid-based map making using particle filters

One of the fundamental tasks in robotics is the creation of a map of the area where the robot is moving. The created map is used for its navigation in this environment. This task is called Simultaneous Localisation and Mapping (SLAM). There are many various algorithms for accomplishing this task, such as Kalman filters or particle filters. At present, mainly the particle filters are used. Together with grid-based maps, they allow the creation of the most precise and consistent maps.

Possibilities of Creation of Community Genealogical Database with Semantic Information

This paper deals with design of database system for innovative approach for rewriting records from seriel sources, mainly from old church registers. Suitable database together with user friendly GUI for as comfortable rewriting as possible is needed. Since the records will be rewritten by volunteers -- mainly amateur genealogists -- we need some reputation system that determines ability of reading of old handwriting. Last part will be focused on getting semantic information from those records. This will be the most important part of the work that allows getting information like average number of children per pair, average age at death and so on.

Improving upon Package and Food Delivery by Semi-autonomous Tag-along Vehicles

This paper aims to improve current last mile distribution and package delivery by introducing basic concept of delivery using Semi-autonomous Tag-along Vehicles (SaTaVs) driven by their own agency and desires. SaTaVs are introduced as vehicles capable of traveling by following leading vehicle and thus reducing requirements for their autonomy while maintaining most of the advantages. Whole system is designed as maintaining long term equilibrium with agents goal in maximizing future investment.

Towards a User-wheelchair Shared Control Paradigm for Individuals with Severe Motor Impairments

This paper presents a work in progress study of a novel user-wheelchair shared control paradigm for individuals with severe motor impairments, which consists of an optimal distribution between several modes, from full user control up to autonomous driving one. To this end, a C400 Permobil wheelchair has been equipped with a control command communication interface and with a scanning laser and a RGB-D sensors to carry out the automation algorithms that are part of the robot operating system (ROS) framework. Moreover, sensor data fusion for map making based on the Bayesian method is applied to the Xtion Pro Live RGB-D camera and the Hokuyo laser sensor data readings. These latter are interpreted by a probabilistic heuristic model that abstracts the beam into a ray casting to an occupied grid cell. Preliminary pilot tests were performed in two different room shapes. The first one in a two room laboratory with a narrow doorway, and the second one in a corridor. The former experiment was dropped due to failure to success, whereas, the latter was a successful one. This has been tested with three different modalities; hand-joystick, tongue-joystick and autonomous modes respectively. The successful results of the second pilot-test have proven the feasibility of using a combination of autonomous and manual control of a powered wheelchair in order to continue development towards a shared-control paradigm.

Overview of bio-inspired control mechanisms for hexapod robot

This paper deals with overview of possible control mechanisms for hexapod robot. Basic characteristics of legged robots, a few existing robots and their pros and cons are described in the introduction of this paper. Main part of the paper is focused on the previous work done in the field of hexapod robot control, especially the usage of evolution techniques like neural networks and genetic algorithms. The last part of this paper is about a hexapod robot of our design.

ROS-based remote controlled robotic arm workcell

This paper describes robotic workplace that is being developed at our faculty. It consists of industrial arm Mitsubishi Melfa RV-6SL, gripper Schunk and six Axis cameras. The purpose of this workplace is to serve students for testing algorithms from artificial intelligence and computer vision. It also gives them the opportunity to work with real industrial manipulator and allow them to test things like kinematics and dynamics of the arm. The arm is inaccessible for students so it can be operated only remotely. The software needed for remote controlling and programming is described in this paper.

Path planning and traversable area marking for stereo vision-based 3D map building

This paper deals with the path-planning of a robot during the exploration and 3D map building in the unknown environment. The robot uses a stereocamera for significant features detection, then matches them in both images and computes their 3D coordinates. This way the robot incrementally creates the 3D map of its surroundings and tries to explore as much area as possible. This paper describes the way of marking the traversable path in the map created so far and the path planning to the next explored goal point. As the map uses points and triangles to represent shape of the surface, the traversable area marking and the path planning is also done in the map which is represented in the same way.

Incremental creation of a 3D map with a stereocamera

This paper describes a concept for the creation of a 3D model of the environment by a mobile robot. This technique is called SLAM — Simultaneous Localization and Mapping. In this case we use stereocamera to find the significant features by the SURF detector. The SURF detector will also serve for the matching of the correspondent points in the both images. These points will then be connected to the triangular mesh and displayed.