Reference Number: INTAS-97-2028
INTAS logo Intelligent Neural System for Autonomous Control of a Mobile Robot
Research activities
Home | Participans | Objectives | Expected results | Publications
Research activities | Scientific results | Summary of results
Research activities
  • The research performed so far opened up interesting research directions for further improvements, related to low-cost sensorics (low-resolution camers, infra-red sensors, low-cost ultrasonic sensors). Therefore the team asks for an extension of the project of 6 months to be able to include these additional results into the project.
  • The contractors were carried out the following work:
    Fachhochschule Ravensburg/Weingarten, Department of the Computer Science, IAF Group on Autonomous Robot Systems, Weingarten, Germany
    Scientific official responsible for the project: Mr K. Shilling
    • Determination of Sensor Characteristics
    • Motion Control Schemes
    • Simulation Tools
    • Test Vehicle Properties
    • Development of System Architecture Outline
    • System Integration
    • Simulation Tests
    • Transfer to Hardware
    • Test and Evaluation

    Universidad Politecnica Valencia, Department of Systems Eng.. Computers and Control, Valencia, Spain
    Scientific official responsible for the project: Mr P. Albertos
    • Traditional Path Planning Methods
    • Motion Control Schemes
    • Simulation Tools
    • Test Vehicle Properties
    • Development of System Architecture Outline
    • Development of neural networks for planning a path for the robot
    • System Integration
    • Simulation Tests
    • Transfer to Hardware
    • Test and Evaluation

    Brest Polytechnic Institute, Department of Computers, Laboratory of Artificial Neural Networks, Brest, Belarus
    Scientific official responsible for the project: Mr V. Golovko
    • Obstacle Avoidance Approaches
    • Application Functional Requirements
    • Development of System Architecture Outline
    • Trade-off between Options
    • Definition of modules
    • Interface definition
    • Sensor data processing and data fusion for building of the local environment map
    • Neural networks for control of the robot on the basis of data from ultrasonic sensors and infrared scanner
    • Neural networks for building of the Global Route map
    • Neural Network Methods for Optimal Path Planning
    • Neural Networks for unsupervised training of the robot by means of interaction with environment
    • System integration
    • Simulation Tests
    • Transfer to Hardware
    • Test and Evaluation
    • Application Relevance

    National Academy of Science. Institute of Engineering Cybernetics, Laboratory of System Identification, Minsk, Belarus
    Scientific official responsible for the project: Mr R. Sadykhow
    • Determination of Sensor Characteristics
    • Motion Control Schemes
    • Test Vehicle Properties
    • Application Functional Requirement
    • Development of System Architecture Outline
    • Trade-off between Options
    • Image processing
    • Development of approaches for movement control of the robot on the basis of information from the video camera
    • Development of neural networks for generation of a route map during the robot movement to the target in an unknown environment
    • System Integration
    • Simulation Tests
    • Transfer to Hardware
    • Test and Evaluation

Designed by Oleg Ignatiuk
The Laboratory of Artificial Neural Networks
Brest State Technical University, Belarus