An Evolutionary Robotics Approach to the Control of Plant Growth and Motion: Modeling Plants and Crossing the Reality Gap

The self-organizing bio-hybrid collaboration ofrobots and natural plants allows for a variety of interestingapplications. As an example we investigate how robots can beused to control the growth and motion of a natural plant, using LEDs to provide stimuli. We follow an evolutionaryrobotics approach...

Full description

Saved in:
Bibliographic Details
Main Authors: Wahby, Mostafa, Hofstadler, Daniel Nicolas, Heinrich, Mary Katherine, Zahadat, Payam, Hamann, Heiko
Format: Conference Proceeding
Language:English
Subjects:
bio-hybrid
Biological system modeling
Data models
evolutionary robotics
Light emitting diodes
Light sources
plant modeling
reality gap
Robot sensing systems
self-organization
Strips
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The self-organizing bio-hybrid collaboration ofrobots and natural plants allows for a variety of interestingapplications. As an example we investigate how robots can beused to control the growth and motion of a natural plant, using LEDs to provide stimuli. We follow an evolutionaryrobotics approach where task performance is determined bymonitoring the plant's reaction. First, we do initial plantexperiments with simple, predetermined controllers. Then weuse image sampling data as a model of the dynamics ofthe plant tip xy position. Second, we use this approach toevolve robot controllers in simulation. The task is to makethe plant approach three predetermined, distinct points in anxy-plane. Finally, we test the evolved controllers in real plantexperiments and find that we cross the reality gap successfully. We shortly describe how we have extended from plant tipto many points on the plant, for a model of the plant stemdynamics. Future work will extend to two-axes image samplingfor a 3-d approach.
ISSN:1949-3681
DOI:10.1109/SASO.2016.8