Producing visually attractive television coverage of sports events and entertainment shows is a challenging business. Today's method of setting up unwieldy cameras on fixed stands or using shoulder-mounted cameras with one cameraman for each camera is costly. The thesis will present an easy to use, effective, and currently unavailable alternative: an autonomously flying micro robot with integrated TV camera.
My objectives of this research are to
show a feasible schedule for developing such a device, ranging from a simple autonomously free flying micro platform (Phase 1), to a hyper intelligent micro robot which has a highly developed ethical awareness, and the ability to carry out autonomously verbally formulated complex missions, to adapt to its environment, and to repair itself (Phase 8);
implement and realize the first three phases of this schedule:
Phase 1: Building a small (< 10 in.) and quiet (< 70 dBA) entity, which is able to stay still in the air, to stabilize itself automatically, and to move in three-dimensional space
Phase 2: Mounting a wireless video camera for conveying pictures from otherwise impossible camera angles: extreme close-ups (distance < 2 in.) and fast camera viewpoint changes (e.g., 20 feet vertical travelling in < 2s) without view obstructing stands, cranes, or dollies
Phase 3: Adding the ability to listen to simple verbal commands like Go up! Turn left! Zoom in! ;
show the possibilities and limitations of speech recognition in this project.
For this Master's thesis research, I plan to realize the following phases:
Phase 1a: testing and evaluating Keyence's Gyrosaucer II E-570.
Phase 1b: building simple FFMP with exobrain and umbilical cord
The mechanically simplest construction, which enables automatic hovering. It consists of:
Phase 2: mounting video camera and wireless transmitter.
There is a variety of commercially available components which have to be evaluated, e.g., Supercircuit's PC-48 (size 1.8 in. square, weight < 1 ounce, resolution 330 lines, low light performance 2.0 lux), MicroVideo's DM30A Miniature Dome Camera, Codex's Micro Video Transmitter, or Spymaster's VID1.
Phase 3a: indirect voice control over radio set (walkie-talkie).
The output of the radio set will be connected to the ground station, where standard software like AT&T's Watson or IBM's ViaVoice recognizes the spoken commands.
Phase 3b: direct voice control through on board microphone and audio processing.
A detailed description of the phases is online at http://www.media.mit.edu/~stefanm/FFMP/FFMP_Phases.html.
Here is the schedule for this research project:
|April - May 1998||Literature review|
|June - July 1998||Implementation of Phase 1a (evaluate Gyrosaucer)|
|Aug - Oct 1998||Implementation of Phase 1b (build simple FFMP)|
|Nov 1998||Implementation of Phase 2 (mount video camera and transmitter)|
|Dec 1998 - Jan 1999||Implementation of Phase 3a (indirect voice control)|
|Feb - March 1999||Implementation of Phase 3b (direct voice control)|
|April - May 1999||Thesis writing|
The following references and hyperlinks as well as many more are on-line at: http://www.media.mit.edu/~stefanm/FFMP/FFMP_Phases.html.
U. Baumann, Fliegende Plattform, master's degree thesis, Institut für Automatik, ETH Zürich, Switzerland (1998).
J. Borenstein, The Hoverbot -- An electrically Powered Flying Robot, unpublished white paper draft, The University of Michigan, Ann Arbor, MI (1992).
Gyrosaucer II E-570: http://ns.idanet.co.jp/keyence/english/corporate/hobby/index.html.
H. Kobayashi, Autonomous Indoor Flying Robot Honey: For communication and/or interaction between semiautonomous agents. http://www.ifi.unizh.ch/groups/ailab/people/hiroshi/honey.html (1997).
J. A. Paradiso, "The interactive balloon: Sensing, actuation, and behavior in a common object." IBM Systems Journal 35, Nos 3&4, 473-498 (1996).
C. Verplaetse, "Inertial proprioceptive devices: Self-motion-sensing toys and tools." IBM Systems Journal 35, Nos. 3&4, 639-650 (1996).