Peter received the degrees of B.S. in Computer Science from National Chiao-Tung University, in 2010, Taiwan, M.S. in Computer Science from National Taiwan University, in 2012, and PhD in Electrical Engineering and Computer Science from MIT, in 2018. He worked with Alberto Rodriguez, and John Leonard in mCube lab. He studied physical contact (e.g. pushing objects) to help robots to understand the physical world. His work on state estimation are published in IROS and ICRA: Shape and Pose recovery from Planar Pushing, More than a Million Ways to Be Pushed, and State Estimation for Planar Pushing and Insertion.

Designing robots to improve people's lives has always been his career goal. He worked on perception problems in DARPA Robotics Challenge (2013-2015), and served as perception and software lead in Amazon Robotics Challenge (Second Place 2015, Third Place in 2016, First Place in Stowing in 2017).


2018/06 Co-founded XYZ Robotics and work as CTO to push the boundary of robotics manipulation technology in industrial applications. We are hiring and forming industrial partners. Webpage

2017/09 Submitted our work on realtime state estimation with tactile and visual sensing. Webpage Preprint

2017/09 Gave talks on "Perception for manipulation" at Foxconn, Dorabot, NCTU ECE, and NCTU CS.

2017/08 Got 1st place in stow task in Amazon Robotics Challenge. Sample Video Team website

Interviewed by NHK

2017/03/21 I gave a talk "Perception for manipulation" in MIT Vision Seminar.

2016/10/07 Our pushing dataset paper is nominated for best paper award in IROS 2016.

2016/09 I was the lead TA of 2.12 Intro to Robotics. I revised the lab materials using ROS framework and designed the final project competition "Pokemon Robotics Challenge." Materials are available here. I got a score of 6.9 out of 7 in course evaluation.

2016/09/01 I gave a webinar on MIT APC approach for ABB research worldwide.

2016/08 I co-hosted the Automation for Warehouse Logistics Materials. We discussed both the Amazon Picking Challenge approaches as well as current needs and technology in Warehouse automation. Talks are recorded and published online.

2016/07/02 Got 3rd place in Stowing task in Amazon Picking Challenge.

Pictures and videos on APC

2016/04/30 We collected a large-scale high-fidelity experimental data set of planar pushing.

Link to the data set

2015/08/11 Got second place in MIT Solve-a-Thon - TEAM p-shirt. We proposed a wearable bend with sensors that helps detect pneumonia.

2015/06/30 One IROS 2015 paper accepted. IROS Talk Video Paper

2015/05/28 Got second place in Amazon Picking Challenge.

Pictures and videos on APC