This page is incomplete. Click here for a full listing of recent publications and preprints.

Scene understanding from 3d data This work is a part of the MARS (Mobile Autonomous Robot Software) 2020 program to develop technologies for object referenced navigation in dynamic urban environments. Our early work concerned robust statistical techniques for extracting coherent planar structures from the large, noisy, cluttered and non-uniformly sampled ladar data typically acquired in such scenes. Other problems like object detection and clustering, and impoving the fidelity of local shape descriptors are currently being pursued. Robust Extraction of Multiple Structures from Non-Uniformly Sampled Data R. Unnikrishnan, M. Hebert Proc. IEEE/RSJ Intl. Conference on Intelligent Robots and Systems, 2003 (IROS ‘03) Vol. 2, pp. 1322–1329 [pdf] Toward Generating Labeled Maps from Color and Range Data for Robot Navigation C. Pantofaru, R. Unnikrishnan, M. Hebert Proc. IEEE/RSJ Intl. Conference on Intelligent Robots and Systems, 2003 (IROS ‘03) Vol. 2, pp. 1314–1321 [pdf]

Globally consistent mapping in large cyclic environments While in the MS program, I worked on constrained optimization formulations of the mobile map-building problem, and on automatically extracting the basis constraint equations through a graph reduction of the problem. This work was in collaboration with Alonzo Kelly at the National Robotics Engineering Consortium (NREC). Efficient Construction of Globally Consistent Ladar Maps using Pose Network Topology and Nonlinear Programming A. Kelly, R. Unnikrishnan Proc. 11th International Symposium of Robotics Research (ISRR 2003) (in press)  [pdf] A Constrained Optimization Approach to Globally Consistent Mapping R. Unnikrishnan, A. Kelly Proc. IEEE/RSJ Intl. Conference on Intelligent Robots and Systems, 2002 (IROS ‘02), Vol. 1, October, 2002, pp. 564–569  [pdf] Globally Consistent Mosaicking for Autonomous Visual Navigation R. Unnikrishnan Master’s thesis, tech. report CMU-RI-TR-02-22, Robotics Institute, Carnegie Mellon University, September, 2002  [pdf] Mosaicking Large Cyclic Environments for Visual Navigation in Autonomous Vehicles R. Unnikrishnan, A. Kelly Proc. IEEE International Conference on Robotics and Automation, 2002 (ICRA ‘02), Vol. 4, May, 2002, pp. 4299–4306  [pdf]