Perching Quadrotor

One of the primary uses of small unmanned vehicles is to gather information about an object or an aspect of the environment using onboard sensors. Multirotors are especially attractive for performing observations because htey are highly maneuverable and capable of hovering. However, current battery technology limits the ability of a multirotor to perform extended observations. Most multirotors have flight times of abound 10-15 minutes, after which they need to either be recharged or have their batteries swapped ouf for fresh ones. This is the primary limiting factor for making extended observations using multirotors, but there are some ways around this issue aside from improvements to battery technology.

One way to perform extended observations with a multirotor is to conserve energy by decreasing the amount of observation time that is spent flying or hovering. This can be achieved by landing or perching. Sometimes landing is not an option due to high traffic, a cluttered environment, or the fact that landing on the ground would not provide an appropriate vantage point for the sensors. In such cases, perching may be a safer alternative that provides a better vantage point for observations.

Our gripper design is inspired by bird feet, with segmented "toes" connected by compliant polyurethane joints with (fishing line) tendons running through them. The gripper is actuated by the weight of the quadcopter putting tension on the tendons, which causes the toes to curl around the perch. When tension is released as the aircraft takes off from it's perch and ceases applying force on the tendons, the joints act as springs and extend the toes, releaseing the perch. This way, the aircraft can perch and take off without using any power for the perching mechanism.

The quadcopter guides itself to the perch using Tau theory, or time-to-contact theory, which mimics the way most animals approach points of interest. The basic idea is that the ratio of the distance to the point of interest to the speed of approach is kept constant throughout the approach. This theory is extended to work for the full 6 degree of freedom quadcopter model in our work. Currently the Tau controller works in the high fidelity SIL part of our unmanned systems testbed, and after some small changes we plan to use it in the HIL simulation followed by a full flight test.