Research

Here are some of my research interests:

1. Safe and Precise Landing

Sponsor: NASA , Jet Propulsion Laboratory

Cameras are light, small, passive – and thus energy efficient - sensors that have been successfully used to estimate the motion of a vehicle by tracking features between consecutive sets of images. Recent research on vision-based motion estimation has demonstrated the ability to estimate the linear and angular displacement of a vehicle using a single camera. Although these techniques are fairly accurate for short motions, they are susceptible to certain limitations such as (i) the inability to distinguish between small rotations and small translations, (ii) the loss of track during sudden motions, (iii) the scale ambiguity when the distance to the tracked features is unknown and (iv) the increased computational requirements when an initial estimate of the performed motion is not available. We address these issues by implementing an enhanced Kalman filter-based estimation scheme for combining navigation information from inertial sensors (IMU) with the visual feature-tracking information from a single camera. Velocity and pose estimates from the inertial sensors are used to resolve the small rotation/translation ambiguity in the motion estimates and provide an accurate initial estimate for the feature-tracking algorithm. This in effect significantly reduces the required time for convergence of the feature-tracking algorithm. In addition, when the field of view changes drastically or when a sufficient number of features are not available for reliably estimating displacements, the enhanced estimator is able to continue tracking the motion of the vehicle by relying solely on the signals from the IMU sensors. Finally, by enhancing the filter with visually extracted motion measurements, the short-term biases of the inertial sensors are estimated on-line and therefore the rate of positioning error accumulation is reduced. This filter is used for tracking the position and orientation of an aerial vehicle (autonomous helicopter) while descending and/or hovering above a potential landing site.

April 02, 2005: Maiden Flight of our Model Helicopter

2. Autonomous Robot Stairclimbing

Sponsor: NSF

In Search and Rescue Scenarios, robots can play an important role to access areas in disaster sites where it is too dangerous or even impossible for humans to go. Stairs are typical obstacles rovers have to overcome in indoor environments. Manually navigating a robot on stairs is a very demanding task. It is therefore desirable for the robot to be able to climb stairs autonomously.

We have devised an algorithm that uses an inertial measurement unit (IMU) and information from a camera to autonomously align with the stairs, climb the stairs, automatically correct the heading even when slipping, and automatically stop upon arriving at the top of the staircase, returning the control back to the operator. From the camera images, the algorithm is capable of detecting the boundaries of the stairs and automatically keeps the robot within a safe zone in the middle.

Look at our videos (video1, video2, video3, video4)

The packbot, climbing up stairs in Walter Library

March 22, 2005: NSF/SSR-RC Demonstration of Stairclimbing in Tampa, FL

Joel Hesch, Tassos Mourikis and me

3. Optimal Cooperative Robot Localization

In Cooperation with: University of Toronto, Supaero, University of Stuttgart

For mobile robots, accurate localization is a key issue. Cooperation between multiple robots can, among other benefits, improve their localization capabilities. We investigate the assumption that it is possible to optimize the trajectories of cooperating robots with respect to their localization performance. One approach consists of optimizing entire trajectories for a group of mobile robots that use one another as localization beacons. The cost function we seek to optimize is a measure of localization uncertainty (as opposed to common criteria such as distance travelled or time). Using an estimation-theoretic framework for cooperative localization, we analyze geometric sensing configurations and set up the optimization problem. We then apply a Sequential Quadratic Programming method to compute optimized trajectories for selected scenarios. For some results, see my research thesis, or the paper presented at ICRA 2004.

4. Controlled Lunar Impact Mission for a Small Satellite Using Low Thrust Electric Propulsion

In Cooperation with: University of Stuttgart

The Institute of Space Systems (IRS) at the University of Stuttgart is currently planning a lunar small satellite mission. The satellite will be equipped with a 6 mN and a 100 mN electric propulsion system. At the end of its primary science mission, it will perform a controlled impact-experiment on the lunar surface, including the soft landing of a small surface unit. We are interested in a numerical simulation and optimization of possible impact trajectories starting from the satellite's initial 100 km polar, circular orbit. The perturbing accelerations being in the same order of magnitude as the thrust, we use thrust vector control for efficient orbit manipulation. First results show that an impact using the electrical thrusters is principally feasible. Using the 6 mN thrusters by themselves for the deorbit maneuver is unadvisable due to the long thrust durations and the very low impact angle, making the impact inaccurate and difficult to control. The 100 mN thruster, however, together with an additional solid rocket motor for a final aposelene boost, yields much more favorable impact conditions at the price of a higher subsystem mass. See my Diploma Thesis for more details, or look at the publication at the International Astronautical Congress in Vancouver.

Home | Research | Publications | CV | Classes | Personal | Contact

The views and opinions expressed in this page are strictly those of the page author.
The contents of this page have not been reviewed or approved by the University of Minnesota.