Seminars

Seminars

Seminars are held on Tuesdays at 3:00 pm in 206 Thomas. Please e-mail comments, questions, and requests to be added to the seminar e-mail list to Maria Koeper.

Monday, October 19, 2009 (Thesis Seminar)
10:00 p.m., Room 306 Thomas
Jeremy Ma
Real-time Applications of 3D Object Detection and Tracking
Robot perception is a fundamental aspect of any autonomous system. It gives the robot the capacity to
make sense of vast amounts of data and gain an understanding of the world around it. An active
problem in the area of robot perception is realtime detection and pose estimation of threedimensional
objects. A framework for object detection and tracking utilizing 3D SIFT features is presented.
Geometric object models are learned in short order time via a training phase and realtime capability is made possible by performing SIFT feature extraction within an adaptive region of interest of the camera image. The experimental results obtained by using this method will show the effectiveness of the approach as compared against ground truth measures in realtime. Using that framework as a basis, extensions to two other problems in robot sensing are then considered: (1) sensorplanning for model identification, and (2) sensorplanning for objectsearch. In the former, a novel algorithm for determining the nextbestview for a mobile sensor to identify an unknown 3D object from among a database of known models is presented and tested across two experiments involving real robotic systems. An information theoretic approach is taken to quantify the utility of each potential sensing action. In the latter area, a novel approach is presented that allows an autonomous mobile robot to search for a threedimensional object using an onboard stereo camera sensor mounted on a pantilt head. Search efficiency is realized by the combination of a coarsescale global search coupled with a finescale
local search, guided by a gridbased probability map. Obstacle avoidance during the search is naturally integrated into the method with additional experimental results on a mobile robot presented to
illustrate and validate the proposed search strategy.

November 3, 2009
Serina Diniega, University of Arizona
Dune and Dune Field Morphology
Many dune fields exhibit regular patterns with apparent characteristic size and spacing, while others contain dunes of very disparate size. Although current dune evolution models are able to replicate isolated dune structures through consistent and reasonable parameter manipulation, they do not yield information about what controls and stabilizes dune field morphology -- local conditions such as topography or sand supply, or general dynamic processes such as those relating to saltation or dune collisions. I will present a multi-scale model developed to explore the influence of dune collisions on pattern-formation within dune fields, and demonstrate that both influx dune sizes (those dunes found at the spatial start of the field) and dune collision dynamics (relating to the exchange of sand between interacting dunes) influence the long-term dynamics of a dune field. I will then discuss the implications these results have with regards to observations of terrestrial and martian dunes, and future dune field modeling efforts.

November 10, 2009 (Thesis Seminar)
Noel DuToit
Robot Motion Planning in Dynamic, Cluttered, and Uncertain Environments: the Partially Closed-Loop Receding Horizon Control Approach
This thesis is concerned with robot motion planning in dynamic, cluttered, and uncertain environments. Successful and efficient robot operation in such environments requires reasoning about the future system evolution and the uncertainty associated with obstacles and moving agents in the environment. Current motion planning strategies ignore future information and are limited by the resulting growth of uncertainty as the system is evolved. This thesis presents an approach that accounts for future information gathering (and the quality of that information) in the planning process. The Partially Closed-Loop Receding Horizon Control approach, introduced in this thesis, is based on Dynamic Programming with imperfect state information. Probabilistic collision constraints, due to the need for obstacle avoidance between the robot and obstacles with uncertain locations and geometries, are developed and imposed. By accounting for the anticipated future information, the uncertainty associated with the system evolution is managed, allowing for greater numbers of moving agents and more complex agent behaviors to be handled. Simulation results demonstrate the benefit of the proposed approach over existing approaches in static and dynamic environments. Complex agent behaviors, including multimodal and interactive agent-robot models, are considered.