Tuesday, July 2, 2013

UAV THAT PERCHES ON THE WALL OF YOUR HOUSE SPYING



 2013.07.02 | 15:36:09 
On this page... (hide)
  1. 1. Vision for Perching UAV
  2. 2. Ongoing Work
  3. 3. Publications
  4. 4. Multimedia
  5. 5. What's New?
  6. 6. Links

1.  Vision for Perching UAV

A flock of small, unmanned air vehicles flies quietly into a city, maneuvering among the buildings. They communicate as they search for places to land, not on streets or flat rooftops but on the sides of buildings and under the eaves, where they can cling, bat or insect-like, in safety and obscurity. Upon identifying landing sites, each flier turns toward a wall, executes an intentional stall and, as it begins to fall, attaches itself using feet equipped with miniature spines that engage small asperities on the surface. Using its propeller in combination with its limbs, the flier can creep along the wall and reorient for a better view.... The fliers stay attached for hours or days, consuming little power and emitting no sound as they monitor the area. When finished, they launch themselves with a jump and become airborne again, ready for their next mission. (from Perching Whitepaper)
Small aircraft move quickly and avoid terrain-based obstacles, but use a relatively large amount of power to maintain lift. Perching on a vertical surface creates previously unavailable surveillance and maintenance abilities that include the added benefit of drawing minimum power. Vertical surfaces are abundant in urban environment, are easy to detect and provide a large surface to land on. 

Sometimes the best place to land is on a wall ...

2.  Ongoing Work

The current work focuses on reducing the challenges involved in landing on vertical surfaces. In order to do so, significant efforts have been dedicated to model accurately the airplane and its suspension in order to predict the conditions for successful landing (see our IJRR paper for more details).

Hybrid diagram illustrating the various phases of landing.
More recently, we have been developing techniques to analyze and exploit funnels in dynamical systems in order to allow for aggressive maneuvers and generally, more robust performances. These techniques will allow us to simultaneously optimize the various parts of an hybrid system for robustness, without having to rely on brute-force techniques (i.e. discretization of the operational state space and simulations/experiments) to evaluate its performances. These effort should lead to an airplane that has been optimized and built for perching, that can perform a robust maneuver even under disturbances and can also recover from failures.

Real (left) and estimated (right) ROA for a simple perching model. The estimated ROA has been computed using a formulation (described here) that estimates close to 80% of the real ROA.
Our work on perching is also a good example of "hybrid locomotion" by combining flying and clinging to walls, which brings out the advantages of both systems. We are currently investigating other combination of other locomotion mode: skittering, flying in close proximity of the wall, jumping, gliding, etc.

Takeoff from the wall
Finally, we are interested in ways to change the behavior of a dynamical system by rapidly changing its proprieties/configuration with lightweight and low-power mechanisms. An example of that are the tunable actuators developed in our lab which should allow landing under a larger range of landing conditions.
We are investigating new PerchingControllers using the Gumstix platform.
March 2013: We are developing a number of new mechanisms for dynamic rotorcraft perching on smooth and rough surfaces. This work is part of new project under ARL MAST, in collaboration with Sean Humbert's lab at U. Maryland and a project with Russ Tedrake's lab at MIT.

3.  Publications

(from PerchingPublications)
Lussier-Desbiens, A., Pope, M., Berg, F., Ern Toh, Z., Lee, J. and Cutkosky, M.R., "Efficient Jumpgliding: Theory and Design Considerations," to be presented at IEEE ICRA 2013 (preprint)
Elena Leah Glassman, Alexis Lussier Desbiens, Mark Tobenkin, Mark Cutkosky, and Russ Tedrake. Region of attraction estimation for a perching aircraft: A Lyapunov method exploiting barrier certificates. IEEE International Conference on Robotics and Automation (ICRA), St. Paul, Minnesota, May 14-18, 2012 (preprint)
Alexis Lussier Desbiens, Alan T. Asbeck and Mark R. Cutkosky, "Landing, Perching and Taking Off from Vertical Surfaces", International Journal of Robotics Research, March 2011 vol. 30 no. 3, pp 355-370.
Alexis Lussier Desbiens, Alan Asbeck and Mark R. Cutkosky, "Hybrid Aerial and Scansorial Robotics", ICRA, May 2010, Anchorage, Alaska.
  • ICRA_2010.mp4: Video for ICRA 2010 showing an overview of the perching project, including both landings and takeoffs
Alexis Lussier Desbiens, Alan Asbeck and Mark R. Cutkosky, "Scansorial Landing and Perching", Proc. 14th International Symposium on Robotics Research, September 2009, Lucerne, Switzerland.
Alexis Lussier Desbiens and Mark Cutkosky, "Landing and Perching on Vertical Surfaces with Microspines for Small Unmanned Air Vehicles (UAVs)," presented at UAV'09, June 8-10, Reno, NV and included in Journal of Intelligent and Robotic Systems, Vol. 57, No. 1-4, p. 313-327DOI: 10.1007/s10846-009-9377-z
Alexis Lussier-Desbiens and Mark Cutkosky, October 2008 whitepaper on our perching approach for small airplanes.

4.  Multimedia

(from PerchingVideos)
  • Video overview of the perching project made for ICRA 2010, including both landings and takeoffs 
  • A 1st rough take-off in Stanford Quad: using a modified version of the airplane
  • Thrown glider landing and perching Filmed Dec. 2008, Edited June 2009 for UAV'09 (89MB Quicktime)
    • Foam core glider is thrown at a rough concrete wall. Ultrasonic sensor initiates pitch-up maneuver at ~5m distance. Landing occurs while velocity is partly forward and partly vertical, between 1-3 m/s. Success was achieved for 30/40 throws, albeit under fairly calm conditions. For engaging and gripping the surface, we use a nonlinear suspension and a special adaptation of spiny toes from SpinyBot?
  • One of the first image we have of perching. A very basic glider, manually thrown and controlled: 

  • Hovering with the Paparazzi open-source controller. We have achieved stable hovering with our small acrobatic plane: 

5.  What's New?

6.  Links

(from PerchingLinks)

No comments:

Post a Comment