I am a Senior Research Scientist with self-driving car company Argo AI working on the Prediction, Deep Forecasting team. I work on modeling the behavior of relevant actors present in the environment of a self-driving vehicle (e.g., other vehicles, bicyclists, pedestrians, etc.). I create and build deep learning models to predict the future trajectories and other useful intentions of such relevant agents (e.g., yielding / non-yielding behavior).

Previously, I spent three years at Motional (formerly nuTonomy) on the Prediction and Behavior Modeling team, working on a similar set of challenges.

Prior to joining industry, I completed my Ph.D. in robotics at the Social Robotics Lab within the Department of Computer Science at Yale University. My advisor was Brian Scassellati.

During my Ph.D., I was interested in machine learning techniques applied to robotics, including hidden Markov models (HMMs) and reinforcement learning. I explored these in the context of human-robot collaboration, with the aim of developing robots that provide supportive behaviors to a person during a physical task. In this scenario, I regarded the human as an agent in the system, and the robot as a learner that needs to adapt to both the task and the human agent.

During the summer of 2017, I had a great time interning at Uber Advanced Technologies Group, in San Francisco. I worked on integrating temporal context into deep learning models for self-driving car perception.

Prior to my time at Yale, I had completed a Master of Engineering from the University of Bristol. Take a look at my resume for more details. A long-form, academic-style CV is also available.

Argo AI

Motional

nuTonomy

Uber ATG

Yale University

Yale Engineering

University of Bristol

University of California, San Diego

My passion in research is to apply machine learning techniques to the field of robotics in order to create autonomous robots / vehicles capable of learning useful behaviors in dynamic environments. Whether applied to physical interactions between robots and people in human-robot collaboration scenarios, or to self-driving cars in complex environments where the autonomous vehicle needs to interact with human drivers on the road, we need to come up with dependable ways of applying learning techniques that make it possible for the robot to learn from both the people it interacts with and its environment.

Recently, I have been focusing on modeling the behavior of the agents that an autonomous vehicle interacts with on the road. Specifically, I have been looking at applying deep learning (including recurrent neural networks) to datasets involving temporal data about the agents present in a self-driving car's environment. Having high-quality predictions of intentions and future trajectories for these agents is crucial if we are to enable robust functioning of self-driving cars.

Here is my Ph.D. thesis: Learning Supportive Behaviors for Adaptive Robots in Human-Robot Collaboration.

Here is a list of my publications, in reverse chronological order (18 - 1):

• Covernet: Multimodal Behavior Prediction using Trajectory Sets [PDF]

  Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2020.

  T. Phan-Minh, E.C. Grigore, F. A. Boulton, O. Beijbom, and E. M. Wolff

• Motion Prediction using Trajectory Sets and Self-Driving Domain Knowledge [PDF]

  Under review.

  F. A. Boulton, E.C. Grigore, and E. M. Wolff

• Preference-Based Assistance Prediction for Human-Robot Collaboration Tasks [PDF]

  Proceedings of the 31st IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2018.

  E.C. Grigore, A. Roncone, O. Mangin, and B. Scassellati

• Predicting Supportive Behaviors for Human-Robot Collaboration [PDF]

  Proceedings of the 17th International Conference on Autonomous Agents and Multiagent Systems (AAMAS) 2018, Extended abstract.

  E.C. Grigore, O. Mangin, A. Roncone, and B. Scassellati

• Discovering the Granularity of Primitive Actions from Human Motion Data in Human-Robot Teaming [PDF]

  Robotics: Science and Systems (RSS) 2017.

  E.C. Grigore and B. Scassellati

• Hierarchical Multi-Agent Reinforcement Learning through Communicative Actions for Human-Robot Collaboration [PDF]

  Proceedings of the Future of Interactive Learning Machines (FILM) Workshop at the 30th Annual Conference on Neural Information Processing Systems (NIPS) 2016, Full paper.

  E.C. Grigore and B. Scassellati

• Comparing Ways to Trigger Migration between a Robot and a Virtually Embodied Character [PDF]

  Proceedings of the 8th International Conference on Social Robotics (ICSR) 2016. Best student paper finalist.

  E.C. Grigore, A. Pereira, J. J. Yang, I. Zhou, D. Wang, and B. Scassellati

• Talk to Me: Verbal communication improves perceptions of friendship and social presence in human-robot interaction [PDF]

  Proceedings of the 16th International Conferences on Intelligent Virtual Agents (IVA) 2016. Best paper finalist.

  E.C. Grigore, A. Pereira, I. Zhou, D. Wang, and B. Scassellati

• Prior Behavior Impacts Human Mimicry of Robots [PDF]

  Proceedings of the 25th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN) 2016.

  A. Suman, R. Marvin, E.C. Grigore, H. Admoni, and B. Scassellati

• Constructing Policies for Supportive Behaviors and Communicative Actions in Human-Robot Teaming [PDF]

  Proceedings of the HRI Pioneers Workshop at the 11th ACM/IEEE International Conference on Human-Robot Interaction (HRI) 2016.

  E.C. Grigore and B. Scassellati

• Modeling Motivational States in Adaptive Robot Companions [PDF]

  AAAI Fall Symposium Series 2015.

  E.C. Grigore, A. Pereira, and B. Scassellati

• Modeling Motivational States through Interpreting Physical Activity Data for Adaptive Robot Companions [PDF]

  Proceedings of the 23rd International Conference on User Modelling, Adaptation and Personalization (UMAP) 2015.

  E.C. Grigore

• Maintaining Engagement in Shared Goals with a Personal Robot Companion through Motivational State Modeling [PDF]

  Proceedings of the Human-Robot Teaming Workshop at the 10th ACM/IEEE International Conference on Human-Robot Interaction (HRI) 2015.

  E.C. Grigore and B. Scassellati

• A Developmentally Inspired Transfer Learning Approach for Predicting Skill Durations [PDF]

  Proceedings of the 4th joint IEEE International Conference on Development and Learning and on Epigenetic Robotics (ICDL-EPIROB) 2014.

  B. Hayes, E.C. Grigore, A. Litoiu, A. Ramachandran, and B. Scassellati

• How to Train Your DragonBot: Socially Assistive Robots for Teaching Children About Nutrition Through Play [PDF]

  Proceedings of the 23rd IEEE Robot and Human Interactive Communication, 2014 (ROMAN) 2014.

  E. Short, K. Swift-Spong, J. Greczek, A. Ramachandran, A. Litoiu, E.C. Grigore, D. Feil-Seifer, S. Shuster, J.J. Lee, S. Huang, S. Levonisova, S. Litz, J. Li, G. Ragusa, D. Spruijt-Metz, M.J. Mataric, B. Scassellati, et al.

• Feasibility of SAR Approaches - Helping Children with Learning Tasks [PDF]

  Proceedings of the International Workshop on Developmental Social Robotics (DevSor): Reasoning about Human, Perspective, Affordances and Effort for Socially Situated Robots at the 26th IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2013.

  E.C. Grigore and B. Scassellati

• Joint Action Understanding Improves Robot-to-Human Object Handover [PDF]

  Proceedings of the 26th IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) 2013.

  E.C. Grigore, K. Eder, A.G. Pipe, C. Melhuish, and U. Leonards

• Towards Safe Human-Robot Interaction [PDF]

  Proceedings of the 12th Annual Towards Autonomous Robotic Systems (TAROS) 2011.

  E.C. Grigore, K. Eder, A. Lenz, S. Skachek, A.G. Pipe, and C. Melhuish

As a graduate student at Yale, I have assisted in teaching the following courses:

Other teaching and mentoring:

• Over the years, I have been mentoring both high-school and undergraduate students who are interested in joining our lab and working with me on different components of my research projects. (2013-present)
• In college, I was a mathematics student-teacher as part of the Students Associates Scheme at Sydney Stringer Academy, UK. (2013)

Throughout my time at Yale, I have been involved with several outreach activities:

• We joined the New York City Science Festival in 2014.
• We hold lab open houses every year, so be on the lookout for news on the lab webpage and join us next time!
• We routinely take our robots to showcase them and talk about our research at local schools in CT.

• Romania is my home. I come from a gorgeous town called Piatra Neamț in the country's north-east. The picture to the right shows one of the historical buildings still standing in the city center (a tower raised by Stephen the Great).
• I love to dance! My hobbies include salsa and tango, which keep me active and on beat and through which I have made some dear friends.

I am always happy to talk about research! Contact me at: