Object Placement as

Inverse Motion Planning


In the summer of 2012, I began my first independent research project in the Learning and Intelligent Systems Lab. The task of picking up and placing an object is an integral task in robotic development and has received a lot of attention. However, most the previous work on placing focused on placing a given object in an exact goal location. This approach fails in several situations where the robot may be holding the object poorly or the object may be particularly hard to place. In these cases, there is a collision between the robot arm and the table, and the place is impossible. In our approach, we take advantage of the known physics governing the dynamics of the object’s fall and other obstacles in the environment to, first, choose a reachable release pose to release the object from, and second, use the robot’s other hand to help guide the object’s trajectory as it is released. I published this work to International Conference on Robotics and Automation (ICRA 2013), and attended the conference in Karlsruhe, Germany presenting this work in an interactive poster session. I also presented this work at MIT EECSCon, an undergraduate conference, and NERC. This research was funded by Colin and Erika Angle and iRobot during my senior year, 2012-2013 through the SuperUROP program. This program aims to make research opportunities more available to undergraduates. I received the EECS Research and Innovation Scholar award for this research.



Object Placement as Inverse Motion Planning
Anne Holladay, Jennifer Barry, Leslie Kaelbling, Tomas Lozano-Perez. IEEE International Conference on Robotics and Automation. Karlsruhe, Germany, 2013.

Object Placement as Inverse Motion Planning
Anne Holladay, Jennifer Barry, Leslie Kaelbling, Tomas Lozano-Perez. MIT EECSCon, 2013. Poster Presentation


Object Placement as Inverse Motion Planning
Anne Holladay, Jennifer Barry, Leslie Kaelbling, Tomas Lozano-Perez. Northeast Robotics Colloquium, October 2012. Poster Presentation.


MIT News, Teaching Robots Lateral Thinking

MIT News, MIT Teaches Robots to Adapt

Advisors: Leslie Kaelbling & Tomas Lozano-Perez

MIT CSAIL LIS