Janusz Bedkowski

Semantic Simulation Engine in Mobile Robot Operator Training Tool

In the paper the semantic simulation engine and its role in multi level mobile robot operator training tool is described. Semantic simulation engine provides tools to implement mobile robot simulation based on real data delivered by robot observations in INDOOR environment. It is important to emphasize that real and virtual part of the training system is integrated. The supervision of real objects such as robots is performed by association with its virtual representation in the simulation, therefore events such as object intersection, robot pitch roll are defined. Semantic simulation engine is composed of data registration modules, semantic entities identification modules (data segmentation) and semantic simulation module. The data registration modules delivers 3D point clouds aligned with ICP (Iterative Closest Point) algorithm accelerated by parallel computation to obtain on-line processing. Semantic entities identification modules provide implementation of methods for obtaining semantic entities from robot observations (already registered). Semantic simulation module executes rigid body simulation with predefined simulation events. It is implemented using NVIDIA PhysX engine. The simulation can be easy integrated with real part of the system with an assumption of robust localization of real entities, therefore Augmented Reality capabilities are available.

GPGPU implementation of On-Line point to plane 3D data registration

The paper concerns the result of the implementation of classic point to plane 3D data registration method with an improvement based on GPGPU parallel computation. 3D data is delivered by mobile robot equipped with 3D laser measurement system for INDOOR and OUTDOOR environments. Presented empirical analysis of the implementation shows the On-Line computation capability using modern graphic processor unit NVIDIA GF 580. The implementation is a part of a project “Semantic simulation engine” composed of following modules: data registration module, semantic entities identification module, semantic simulation. The goal of the project is to deliver software tools capable to create virtual model of the environment based on robot observations and perform semantic simulation, where all virtual entities correspond to real one. Possible practical application of the project are supervision and control of robotic system and mobile robot operator training using Augmented Reality techniques. Data registration module is composed of the implementation of point to point and point to plane classic methods improved by the usage of parallel computation. In this paper it has been shown that the implementation of GPGPU point to plane registration method is accurate in INDOOR structured environment but it has some difficulties in accurate alignment in OUTDOOR environment.

RISE mobile robot operator training design

In the paper the framework for RISE (risky intervention and environmental surveillance) mobile robot operator training design is presented. The basic idea of the framework is to provide advanced software tools to improve the performance of the mobile robot simulation design that are applicable for operator training. The simulator is the end product of the framework. To achieve realistic rigid body simulation and realistic rendering the NVIDIA PhysX engine and the OGRE engine are used. It has to be noticed that the compatibility between the framework software and CAD modeling tools such as Solid Works was essential, therefore virtual models are exchangeable via COLLADA files. The simulator can be integrated with control panel of the mobile robot. It is possible to simulate the behavior of several types of robots such as caterpillar or wheeled. It is important to emphasize that presented result is a new approach related to mobile robots applications and potentially can improve the process of advance mobile robot application design and professional training of the operators.

Improvement of the Robotic System for Disaster and Hazardous Threat Management

Abstract Following paper shows an improvement of the robotic system (Maslowski, 2004 and 2006) for disaster and hazardous threat management. The system uses two types of robots for investigating the potential risks in unknown environment – teleoperated and fully autonomous mobile platform. The teleoperated robot equipped with video cameras and 5DOF arm is used for hazardous materials detection and neutralization. The improvement is based on the autonomous mobile robot development based on hybrid composition of 3D laser scanners providing 3D feedback to base station. The idea of real time 3D map building is shown. The new idea of the usage given map to the supervision module of the robotics system is presented.

On-line data registration in OUTDOOR environment

In the paper an algorithm of 3D data registration based on CUDA implementation is shown. The research is related to the problem of collecting 3D data with laser measurement system mounted on rotated head, to be used in mobile robot applications. Assumed performance of data registration algorithm is achieved, therefore it can used as On-line. The ICP (Iterative Closest Point) approach is chosen as registration method. Computation is based on massively parallel architecture of NVIDIA CUDA. The presented concept of 3D data matching is based on parallel computation used for fast nearest neighbor search. Nearest neighbor search procedure is using 3D space decomposition into cubic buckets, therefore the time of matching is deterministic.

The hierarchical knowledge representation for automated reasoning

In the paper the study of knowledge hierarchical representation for automated reasoning is presented. The hierarchical knowledge representation is proposed for predictive modeling purpose. It is improved an effective automated reasoning structure for data set analyzes and making decisions based on complex relations between this data. It is important to emphasize that it is not considered a — priori knowledge concerning data structure, therefore the approach automatically discovers particular constraints between data. It provides a technique of the verification the hierarchical knowledge representation building process that can be useful for the model justification. The presented numerical experiment shows an advantage of proposed approach. It is assumed that the presented automated reasoning can be used for classification purpose where there is a difficulty of proper classifier choice.

Cognitive supervision and control of robotic inspection-intervention system using qualitative spatio-temporal representation and reasoning

In this paper a new concept of Mobile Spatial Assistive Intelligence (MSAI) applied for the control and supervision of web connected mobile robots is introduced. Mobile Intelligence is implemented as MSAS (Mobile Spatial Assistance System) - a mobile robot available to perform Qualitative Spatio-Temporal Representation and Reasoning (QSTRR). Taking up this subject is justified by the need of developing new methods for the control, supervision and integration of existing modules (inspection robots, autonomous robots, mobile base station). The research problem comprises the web connected mobile robots system development structure with the exemplification based on the inspection-intervention system. Using QSTRR for supervision and control of robotics system is a novel approach and I believe that this framework can decrease the cognitive load for a human operator. The main task for MSAS is to build the semantic model of the structured environment based on its observations. This semantic model is used by a qualitative reasoner - an important component of MSAI, to perform spatial entities supervision based on the spatial relations and the spatio-temporal events. The experiments results obtained using the prototype of MSAS in real indoor environment are also discussed. This approach can be used for the design of the intelligent human machine interface that can help in controlling the robot during the risky scenarios.

Advanced computer graphic techniques for laser range finder (LRF) simulation

This paper show an advanced computer graphic techniques for laser range finder (LRF) simulation. The LRF is the common sensor for unmanned ground vehicle, autonomous mobile robot and security applications. The cost of the measurement system is extremely high, therefore the simulation tool is designed. The simulation gives an opportunity to execute algorithm such as the obstacle avoidance[1], slam for robot localization[2], detection of vegetation and water obstacles in surroundings of the robot chassis[3], LRF measurement in crowd of people[1]. The Axis Aligned Bounding Box (AABB) and alternative technique based on CUDA (NVIDIA Compute Unified Device Architecture) is presented.

Efficient defect structure analysis in semi-insulating materials by support vector machine and relevance vector machine

We propose new approach for defect centers parameters extraction in semi-insulating GaAs. The experimental data is obtained by high-resolution photoinduced transient spectroscopy (HR-PITS). Two algorithms have been introduced: support vector machine - sequential minimal optimization (SVM-SMO) and relevance vector machine (RVM). Those methods perform the approximation of the Laplace surface. The advantages of proposed methods are: good accuracy of approximation, low complexity, excellent generalization. We developed SVM-RVM-PITS system, which enables graphical representation of Laplace surface, defining local area for defect parameter extraction, choosing the SVM or RVM method for approximation, calculation of the Arrhenius line factors and finally the parameters of the defect centers.

Recognition of defect structure of Si(A4) by on-line support vector machine

In this paper the application of on-line support vector machine to spectral surface approximation is presented. The experimental data were obtained by the photocurrent decay measurement as function of time and temperature for a sample of neutron irradiated silicon. This approach enabled to extract the deep level center defect parameters: activation energy and pre-exponential factor.