X-Hab tele-operated greenhouse

X-Hab tele-operated greenhouse

Our long-term goal is to develop an autonomous green house in which all tasks are performed automatically.  Water and nutrients will be delivered locally on-demand and fruit will be harvested optimally, that is when they are ripe and when its removal is best for the plant. Plants will drive the robots’ activities in the garden using sensors to monitor their local environment conditions, a plant-specific model of growth for making predictions about the state of fruit, and interaction with robots for establishing an inventory of fruit.

The current focus of the project is on tele-operation. The rational is simple: showing that we can tele-operate it demonstrates that the system fulfills the minimal mechanical and sensorial requirements, which can then serve as a baseline for autonomous operation. We have also begun to create the data basis for (un)supervised machine learning of plant status and manipulation planning by developing a 2D gantry system that collects RGB-D readings from Strawberry plants.

The "Datagarden" (left), 3D reconstruction of Strawberry plants (middle), labeled image using "LabelMe"

The “Datagarden” (left), 3D reconstruction of Strawberry plants (middle), labeled image using “LabelMe”

History

Prototype Robotic gardening System

This project started as an undergraduate project course 6.084/086 taught at MIT during Fall 2008. The project was framed as addressing a grand challenge: to create a robotic gardening system. Solving the grand challenge required designing and programming robots to interact effectively and autonomously with the real world. We developed the class hardware infrastructure consisting of six robots with an iCreate base and a 4DOF arm with eye-in-hand configuration and an optional watering system and four cherry tomato plants, each with its own local sensing and computation packaged in an embedded computer. The robots and plants were networked together as a mesh network. The plants have the ability to monitor their soil humidity and issue watering requests. They also have the ability to database the location and color-level of the tomatoes. The robots have the ability to visit a specific plant to deliver water or to locate and grasp a tomato. Users have the ability to request tomatoes for salad. In response to user requests, the system decides which specific plants have the ripest tomatoes and assign parallel harvesting tasks to robots.

This preliminary work has led to a UROP project at CU Boulder (autoponics.org), sponsoring by NASA’s x-hab competition with the goal to develop a tele-operated greenhouse in the academic year 2012/2013 as well as a NASA Early Career grant to develop the fundamental perception and manipulation capabilities future autonomous systems will need. We have also begun studying human factors, i.e., the influence of growing plants in confined environments. More specifically, we are interested into studying, which tasks contribute to astronaut’s well-being and which tasks need to be automatized.

Team

  • David Coleman
  • Nicholas Farrow
  • Heather Hava
  • Daniel Zukowski
  • Aerospace Graduate design project class Fall 2012/Spring 2013
  • Advanced Robotics class Spring 2013

Publications

N. Correll, N. Arechiga, A. Bolger, M. Bollini, B. Charrow, A. Clayton, F. Dominguez, K. Donahue, S. Dyar, L. Johnson, H. Liu, A. Patrikalakis, T. Robertson, J. Smith, D. Soltero, M. Tanner, L. White, D. Rus. Indoor Robot Gardening: Design and Implementation. Intelligent Service Robotics. Special Issue on Robotics in Agriculture.

N. Correll and D. Rus. Peer-to-Peer Learning in Robotics Education:  Lessons from a Challenge Project Class. ASEE Computers in Education Journal. Special Issue on Novel Approaches in Robotics Education. To appear. [preprint]

N. Correll, N. Arechiga, A. Bolger, M. Bollini, B. Charrow, A. Clayton, F. Dominguez, K. Donahue, S. Dyar, L. Johnson, H. Liu, A. Patrikalakis, T. Robertson, J. Smith, D. Soltero, M. Tanner, L. White, D. Rus. Building a Distributed Robot Garden. In IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 1509-1516, St. Louis, MO. Finalist for NTF Best Paper Award.

Funding

  • NASA Early Career Grant
  • NASA Space Technology Research Fellowship (Heather Hava)
  • NASA x-hab competition

Media Coverage