Speaker:

Mr. Ibrahim Kabir

Graduate Student

Control and Instrumentation Engineering Department

King Fahd University of Petroleum and Minerals

Dhahran, Saudi Arabia

Abstract:

Robot navigation has seen many advances, especially since robots were introduced into shared spaces with humans. This has made it necessary for robots to be socially aware and behave in a compliant and non-threatening manner in the presence of humans when carrying out their tasks. This problem increases when robots navigate in crowds, but researchers have used Deep Reinforcement Learning (DRL) Techniques to learn socially cooperative policies in the presence of multiple humans. In this work, a Crowd-Robot Interaction (CRI) framework is modelled. Additionally, pairwise interactions are modelled with a self-attention mechanism, and Human-Robot, as well as Human-Human interactions, are modelled within the deep reinforcement learning framework. Various experiments demonstrate the effectiveness and efficiency of the model when navigating in dense crowds.

Speaker Bio :

Mr. Ibrahim Kabir received his Bachelor’s degree in Mechatronics Engineering from Bayero University, Kano in 2021. He is currently pursuing his M.Sc. degree in Systems and Control Engineering in the Control and Instrumentation Engineering Department at King Fahd University of Petroleum & Minerals, Saudi Arabia. His Areas of Interest include Socially-Aware Mobile Robot Navigation, Collaborative and Multi-Robot Navigation, and Renewable Energy.