Farshad Arvin

Development of a Miniature Robot for Swarm Robotic Application

Biological swarm is a fascinating behavior of nature that has been successfully applied to solve human problem especially for robotics application. The high economical cost and large area required to execute swarm robotics scenarios does not permit experimentation with real robot. Model and simulation of the mass number of these robots are extremely complex and often inaccurate. This paper describes the design decision and presents the development of an autonomous miniature mobile-robot (AMiR) for swarm robotics research and education. The large number of robot in these systems allows designing an individual AMiR unit with simple perception and mobile abilities. Hence a large number of robots can be easily and economically feasible to be replicated. AMiR has been designed as a complete platform with supporting software development tools for robotics education and researches in the Department of Computer and Communication Systems Engineering, UPM. The experimental results demonstrate the feasibility of using this robot to implement swarm robotic applications.

A Short-Range Infrared Communication for Swarm Mobile Robots

This paper presents another short-range communication technique suitable for swarm mobile robots application. Infrared is used for transmitting and receiving data packets and obstacle detection. The infrared communication system is used for an autonomous mobile robot (UPM-AMR) that will be used as a low-cost platform for robotics research. A pulse-code modulation (PCM) digital scheme is used for transmitting data. The reflected infrared signal is also used for distance estimation for obstacle avoidance. Analysis of robot’s behaviors shows the feasibility of using infrared signals to obtain a reliable local communication between swarm mobile robots.

Swarm Robots Long Term Autonomy Using Moveable Charger

This paper proposes an alternative docking charger station for mobile robots. One crucial task in swarm robots scenario is to find low battery robots and recover them. Overcoming this problem requires a versatile mobile charging station that is able to actively locate and charge inactive mobile robots unlike the conventional stationary docking stations. The mobile charger robot should perform independent tasks other than the swarm robots global task. It uses similar communication method with other swarm robots, however new preprogrammed instructions are required to facilitate battery charging task for swarm robots.

Design of a differential-drive wheeled robot controller with pulse-width modulation

This paper presents mobile robots motion control technique based on pulse-width modulation (PWM). This technique is employed on AMiR which is an autonomous miniature robot for swarm robotic platform. That uses differential drive with a caster wheel configuration. Robots motors enable to work with different speed in different direction, forward and reverse. A microcontroller as the main processor is deployed to generate motor control pulses and manage duty-cycle of PWM signals. Two methods in robot trajectory control which are rotation and straight movement are described in this paper. Time estimation and also speed selection calculations illustrate the feasibility of this technique to be used in mobile robot motion control problem.

Implementation of a Cue-Based Aggregation with a Swarm Robotic System

This paper presents an aggregation behavior using a robot swarm. Swarm robotics takes inspiration from behaviors of social insects. BEECLUST is an aggregation control that is inspired from thermotactic behavior of young honeybees in producing clusters. In this study, aggregation method is implemented with a modification on original BEECLUST. Both aggregations are performed using real and simulated robots. We aim to demonstrate that, a simple change in control of individual robots results in significant changes in collective behavior of the swarm. In addition, the behavior of the swarm is modeled by a macroscopic modeling based on a probability control. The presented model in this study could depict the behavior of swarm throughout the performed scenarios with real and simulated robots.

A Real-Time Note Transcription Technique Using Static and Dynamic Window Sizes

This paper presents a real-time signal processing technique using a hardware interface based on the microcontroller to process audio music signals to standard MIDI data. A technique for transcribing music signals by extracting note parameters is described. Two different approaches using static and dynamic window sizes to convert the voice samples for real-time processing without complex calculations are proposed. The transcribed data generated shows the feasibility of using microcontrollers for real-time MIDI generation hardware interface.

An Approach for Heartbeat Sound Transcription

This paper presents a heart sound transcription technique to provide semi-structured musical data. The heart beat samples were captured using stethoscope and recorded in pulse-code modulation (PCM) method into several differing audio file formats. Pitch tracking algorithm were utilized for the heartbeat transcription into MIDI generation. We propose the use of several sound processing techniques to improve the transcription. Several experiments using a collection recorded heartbeats illustrate the feasibility of using this note transcription approach in heart sound.