This site will work and look better in a browser that supports web standards, but it is accessible to any browser or Internet device.
My research is mainly focused on state estimation, data fusion, path planning and control for small unmanned aerial vehicles (UAVs, commonly called drones in the press). The drivers of this research have been two areas critical to improving vehicle capabilities: perception (transforming sensor data into knowledge of the vehicle state and its surroundings) and persistence (exploiting energy available in the environment to improve range and endurance). Problems in state estimation and planning are strongly coupled in both areas: with limited sensors on-board the vehicle, the flight path has a significant effect on the quality of information gained about the environment.
Perception research has focused on obstacle mapping and vehicle state estimation using monocular cameras combined with inertial sensors. These maps can the be used to plan safe trajectories through the environment. The ultimate goal is to enable a small autonomous helicopter to fly through a complex, unstructured environment such as a forest or an urban area.
Research in persistence has been driven by the limited range and endurance inherent to small UAVs: fuel (or on-board energy stored in batteries) is limited and performance is limited by the physics governing the vehicle. However, large birds (which are similar in scale to small UAVs) have evolved flight techniques that exploit energy available in the atmosphere to fly hundreds of kilometers without flapping. In effect, intelligence compensates for physical limitations. Inspired by birds and human sailplane pilots, my students and I are developing algorithms that enable small UAVs to soar.
Combining the improved perception provided by novel sensing and processing systems with the improved persistence provided by autonomous soaring will enable long-duration missions that are far beyond the capabilities of today's robotic aircraft. Eventually a soaring capable small UAV will be able to follow a migrating bird, providing close-up in-flight video as well as detailed measurements of the atmospher through which the bird is flying. Successful completion of such a challenging mission will show that robotic aircraft can finally rival birds in capability.
My research is funded by the National Science Foundation, the Office of Naval Research, the Vertical Lift Consortium, and through the Penn State Vertical Lift Research Center of Excellence.
For details of the research undertaken by me and my students please see the webpage of the Air Vehicle Intelligence and Autonomy lab: AVIA