Peter Tamas Szemes

Pedestrian-behavior-based mobile agent control in intelligent space

This paper investigates a human walking behavior-based mobile robot control system. The Intelligent Space (iSpace) is a distributed sensory system, which is the background infrastructure to observe human walking in a limited area. The observation of human walking behavior is applied to train fuzzy-neural networks (FNN). The trained FNNs are applied to approximate the obstacle avoidance behavior of human walking. The paper introduces the iSpace and the mobile agents, which are mobile robots, utilizing the intelligence of the iSpace. The observed and trained human walking behaviors are applied to control the mobile agent in a human-robot shared environment. Experimental results demonstrate the effectiveness of the FNN-based control system.

Development of a single-master multi-slave tele-micromanipulation system

This paper presents a novel single-master (PHANToM haptic device) multislave telemicromanipulator system using two parallel mechanism micromanipulators as a slave device. Firstly, the motivation of this research is mentioned. Then, the system structure and the configuration of our experimental system are introduced. In this system, there exists singular position problem which prevents precise and flexible movement of parallel mechanism manipulators. Through several simulations like a rectilinear movement, singular position and the manipulability of the dual-micromanipulator system is analyzed. Also, the different kinematical configuration and the different DOF between the master and the slave introduces a mapping problem which can be serious for some cooperative manipulations. Therefore, the mapping method between the master and the slave is discussed in this paper. A manipulation strategy to build the system that human and both manipulators perform the cooperative manipulation is also introduced. Finally, several experimental and simulation results are shown to verify the proposed system.

Nonlinear disturbance compensation for haptic device

The Sensor Glove II (The SG II) is a glove type haptic device. The SG II supports 20 degrees of freedom which corresponds to the total DOF of the human hand. It can measure the force and the angle of each joint and feedback the force from the manipulated environment in each joint. The motors move pulleys of the device through wires in tubes. The nonlinear friction causes disturbances and difficulties in the control. This paper proposes a control method with disturbance compensation based on a sliding mode observer for SG II. If the system is held on a sliding surface, the system is robust against certain parameter changes and certain nonlinearity. The sliding mode observer is an efficient scheme for compensating nonlinear disturbance. The results of free motion experiments show good performances.

Welding trajectory reconstruction based on the Intelligent Space concept

Superflexible robot programming reduces the profitable batch sizes of industrial robots. The paper presents a new way to determine the trajectory for welding robots based on the Intelligent Space concept. The system uses two cameras and edge detection with other image processing algorithms to find the welding path in the image. The three dimensional trajectory is obtained by stereo vision; afterwards it is transformed to the actual robot language.

Friction compensation for 6DOF Cartesian coordinate haptic interface

In this paper, control schemes for haptic interface for micromanipulation systems is mainly discussed. We have proposed a tele-micromanipulation system, which enables human operators to operate micro tasks, such as assembly or manufacturing, without feeling stress. This paper focuses on the haptic sense, which gives the operators the feeling of presence. The mechanical instrument applied in the human interface device usually has a reasonable immanent friction. This friction must be compensated in a way that the operator should not feel this friction force but the force from the manipulated environment. The main contribution of this paper is a direct model based chattering free sliding mode friction estimator and compensator for a mechanical instrument. Experimental results are presented.

“Energymon”: Development of wireless sensor network based Decision Support System to monitor building energy performance

This paper presents a new technology to monitor and improve building energy performance according to the 2002/91/EC directive. The system includes a ZigBee based wireless sensor system, which is installed in the building to measure temperature, heat flow, boiler performance, gas and electricity consumption. The obtained data is processed by a Decision Support System, which (1) calculates the current energy efficiency, (2) generates reconstruction plans and budget, and (3) calculates return of investment. The optimal (i.e. highest return of investment) reconstruction plan is selected and recommended to the user (decision maker). The optimization of reconstruction strategies (options among reconstruction plans) can be determined by setting the available budgets and other available resources such as renewable energies. The performance of the system was tested on a 3000 m2 school building complex, including of 150-year-old and 10-year-old structures.

Mobile Agent Gain Scheduler Control in Inter-Continental Intelligent Space

Intelligent space (iSpace) is a space (room, corridor, or street), which has distributed sensory intelligence and it is equipped with mobile agents. In this paper new concept of the iSpace to connect distant places, over packet switched network is introduced. Experimental iSpace of Hashimoto Lab, University of Tokyo, Japan and Advanced Diagnosis Automation and Control Lab, North Carolina State University, USA is connected to form an inter-continental intelligent space. Conventionally, in order to control an application over a data network, a specified control of teleoperation algorithm is usually required for controller design to compensate network delay effects. Therefore an existing controller has to be redesigned or replaced by a new controller system. This replacement process is usually costly, inconvenient, and time-consuming. In this paper, a novel methodology to enable existing controllers for networked control and teleoperation by middleware is introduced. Mobile agent located in Japan is controlled by tracking controller located in USA to show the effectiveness of the middleware methodology.

Robot navigation framework based on reinforcement learning for intelligent space

Navigation in an unknown environment is still a hard problem, because mobile robots need topological maps in order to operate in the environment. Building a map of the environment while also using it for learning is of prime importance for mobile robots but until recently, it has only been confined to small-scale environments. This paper describes a mobile robot navigation framework integrated into the intelligent space environment. in the intelligent space, several distributed intelligent network devices communicate and share their information about the environment. In this environment mobile robots can be tracked with ultrasonic positioning system and the topological map can be build using laser range finders.

An evaluation of grasp force control in single-master multi-slave tele-micromanipulation

This paper proposes human-robot cooperative tele-micromanipulation system with novel single-master (PHANToM haptic device) multi-slaves (6 D.O.F parallel manipulator). The different kinematical configuration and the different D.O.F between the master and the slave introduces a mapping problem which can be serious for some cooperative manipulations. The position/force virtual mapping method is implemented in the single-master dual-slave tele-micromanipulation system to realize the human-robot cooperative internal force regulation while grasping task. The generated reference trajectories of both manipulators by the virtual mapping method are shared with the cooperative impedance control of multi-slaves to regulate the internal force while grasping an object. Lastly, an improved internal force regulation using the SMMS (single-master multi-slave) tele-micromanipulation system while grasping an object is verified by the pick-and-place experimental results.

Guiding and communication assistant for disabled in intelligent urban environment

In this paper, the intelligent space concept is applied for helping disabled or blind persons in crowded environments such as train stations, or airports. The main contribution of this paper is a general mathematical (fuzzy-neuro) description of the obstacle avoidance method (walking habit) of moving objects (human beings) in a limited area scanned by the intelligent space. A mobile robot with extended functions is introduced as a mobile haptic interface, which is assisted by the intelligent space. The mobile haptic interface can guide and protect a blind person in a crowded environment with the help of the intelligent space. The intelligent space learns the obstacle avoidance method (walking habit) of dynamic objects (human beings) by tracing their movements and helps to the blind person to avoid the collision. The prototype of the mobile haptic interface and simulations of some basic types of obstacle avoidance method (walking habit) are presented.