Skip to ContentSkip to Navigation
Research Bernoulli Institute Calendar

Colloquium Artificial Intelligence - dr. H.A. Kasaei University of Groningen

When:Tu 08-06-2021 15:00 - 16:00
Where:Online (see below)

Title: Towards lifelong robot-learning in human-centric environments: How robots can adapt to new environments incrementally?



While a lot of developments have been made in the field of robotics, computer vision, and machine learning, service robots do not yet live among humans to assist in various daily tasks. The underlying reason is that robots are usually painstakingly coded and trained in advance to perform object perception and manipulation tasks in the right way. Moreover, they are trained once all data has been gathered, making them strongly dependent on the quality and quantity of training data. Therefore, the knowledge of such robots is fixed after the training phase, and any changes in the environment require complicated, time-consuming, and expensive robot re-programming by expert users.

In human-centric environments, it is not feasible to assume that one can pre-program all necessary object categories and grasp templates for the robots. To operate in such dynamic environments, I believe an appropriate approach is to make robots capable of learning in an open-ended fashion by interacting with non-expert users. In this talk, I will give a brief overview of my research on "Lifelong Robot Learning" before diving into one of my latest research on “Simultaneous Multi-View Object Grasping and Recognition in Open-Ended Domains”. In this project, “open-ended” implies that the set of object categories to be learned is not completely known in advance. The training instances are extracted from the online experiences of a robot and become gradually available over time. Moreover, apart from robot self-learning, non-expert human users could interactively guide the process of experience acquisition by teaching new concepts, or by correcting insufficient or erroneous concepts. This way, the robot will constantly learn how to help humans in everyday tasks by gaining more and more experiences without the need for re-programming.

Short bio:

Hamidreza Kasaei is an assistant professor of robotics in the Department of Artificial Intelligence at the University of Groningen, the Netherlands. He received a Ph.D. degree in robotics from the University of Porto, Portugal, in 2018. He was a visiting researcher at Imperial College London, UK, in 2016. Kasaei’s research interests focus on the intersection of robotics, machine learning, and machine vision. His main research goal is to achieve a breakthrough by enabling robots to incrementally learn from past experiences and safely interact with human users based on lifelong machine learning techniques. He has been serving as Associate Editor for IEEE International Conference on Robotics and Automation (ICRA) since 2019, and for IEEE International Conference on Intelligent Robots and Systems (IROS), 2021.