Demonstration-Based Initialization of Reinforcement Learning Algorithms for Efficient Robotic Control

Kindred’s Sort product is a robotic system that operates in e-commerce distribution centers to sort and handle apparel and general merchandise. The deployed system is controlled through a combination of artificial intelligence and human-in-the-loop teleoperation. The proposed project involves applying techniques from artificial intelligence (specifically machine learning and reinforcement learning) to improve the ratio of automatic control to human control. The core hypothesis of the project is that historical data collected from human teleoperation of the robots performing object-grasping tasks can be used to train the robots to pick up items automatically. This task is a challenging research problem at the cutting edge of robotic control and AI, and it will be tackled with a combination of state-of-the-art academic research and internally-developed algorithms.

Faculty Supervisor:

Sven Dickinson

Student:

Partner:

Kindred AI

Discipline:

Computer science

Sector:

Technology; Commercial Services; Advanced Manufacturing

University:

University of Toronto

Program: