Autonomous Mobile Robots Searching Methods By: Alex Morales - PowerPoint PPT Presentation

Autonomous Mobile Robots Searching Methods By: Alex Morales Graduate Mentor : Joey Durham Faculty Advisor: Francesco Bullo

Better Algorithms  Design algorithms for better multi-robot coordination  Design a Search Algorithms for robots Better understand robot – surrounding − interaction

Autonomous Robots ● What are autonomous robots? ● Why should we study them? ● How will they help us? Google images

Applications  Make human life easier – Perform common tasks, help the handicapped  Search and rescue robos Google images

Project Goals  Build a foundation for searching algorithms for autonomous robots  Create an efficient algorithm to search for an object in a known map. – Minimize time

Project Goals Target object Static Environment Obstacle/Wall Robot

Project Goals Target object Static Environment Obstacle/Wall Robots

Project Goals Target object Static Environment Obstacle/Wall Robots

Project Goals Target object Unknown Static Environment Robot

Project Goals Target object = ? Unknown Environment Robots

Equipment: Robot hardware Stereo Vision Camera On board computer Laser range- finder Joey Durham

Player/Stage Interface  Is a open source software that enables us to do research.  Player is a network server for robot controls  Stage simulates mobile robots in 2-D.  Advantages vs "All the world's a stage, And all the men Disadvantages and women merely players." - William Shakespeare, As You Like It Joey Durham

Searching Algorithm Receive laser data Check if  Main part of algorithm object was found – Design, Test, Refine Create frontier Decide next move Object found finish.

Design • Check for Frontier • Visibility of the robot • Polygons • Save polygons instead of having to save every point of the frontier

Determining how to move ● Default left ● Decision based on Frontier

Result so far ...

Result so far ...

Result so far ...

Result so far ...

Result so far ...

Exploration Graph • Map of the Robot • Graph of nodes/positions of robots

Conclusion • Field of Vision of Robot ● Visible polygon • Decision making part of algorithm • Based on how frontier/boundary are created and stored – Intersections merge frontier arcs. ● Navigate through map with an exploration graph.

Acknowledgements ● Mentor Joey Durham ● Dr. Francesco Bullo ● Ofelia Aguirre ● Eric Pressly ● CNSI ● NSF ● SIMS and EPSEM programs

Multiple Robots Object 2

Laser Rangefinder ● Model: Hokuyo URG ● Wavelength: 785 nm ● Range: 4 meters ● Scan rate: 10 Hz ● Resolution: 0.36 degrees ● Price: $2500 Source: http://www.videredesign.com/robots/era_mobi.html

Robot Chassis ● Model: Videre ERA-MOBI ● Size: cm(L) x 37cm (W) x 18 cm(H) ● Batteries: 3x 12V 7 Amp-Hour (4-5 Hours) ● Encoder accuracy: 500 counts/rev ● Speed: Up to 2.0 m/s ● Capacity: 20 kg (44 lbs) ● Price: $2350 Source: http://www.videredesign.com/robots/era_mobi.html

Conclusion Object 1