Shafeef Omar

I am a research fellow with the Dynamic Legged Systems Group, headed by Claudio Semini at the Istituto Italiano di Tecnologia. Previously, I was a Research Associate at the Indian Institute of Science , working on Continual Learning at the Artificial Intelligence and Robotics Laboratory. I graduated from the Indian Institute of Technology Madras (IIT-M) with a Dual Degree, Bachelor’s and Master’s degree, in Engineering Design.

In my current research, I focus on inducing biases using model-based priors to learn safe and robust vision-aided locomotion. I am broadly interested in supervised learning, reinforcement learning and control theory to apply them in robotics for control and planning in ardous scenarios.

news

Sep 30, 2023	SafeSteps has been accepted to the IEEE-RAS International Conference on Humanoid Robots 2023.
Jul 24, 2023	The preprint of our latest paper on Safe Foothold Adaptations for Legged Robots, SafeSteps, is now available. See the publications section.
May 22, 2022	Joined the Dynamic Legged Systems Lab (DLS@IIT).

selected publications

2023

SafeSteps: Learning Safer Footstep Planning Policies for Legged Robots via Model-Based Priors

Shafeef Omar, Lorenzo Amatucci, Giulio Turrisi, Victor. Barasuol, and Claudio Semini

In IEEE-RAS International Conference on Humanoid Robots, 2023

Abs arXiv Bib PDF Website

We present a footstep planning policy for quadrupedal locomotion that is able to directly take into consideration a-priori safety information in its decisions. At its core, a learning process analyzes terrain patches, classifying each landing location by its kinematic feasibility, shin collision, and terrain roughness. This information is then encoded into a small vector representation and passed as an additional state to the footstep planning policy, which furthermore proposes only safe footstep location by applying a masked variant of the Proximal Policy Optimization (PPO) algorithm [1]. The performance of the proposed approach is shown by comparative simulations on an electric quadruped robot walking in different rough terrain scenarios. We show that violations of the above safety conditions are greatly reduced both during training and the successive deployment of the policy, resulting in an inherently safer footstep planner. Furthermore, we show how, as a byproduct, fewer reward terms are needed to shape the behavior of the policy, which in return is able to achieve both better final performances and sample efficiency.
@inproceedings{omar23humanoids, author = {Omar, Shafeef and Amatucci, Lorenzo and Turrisi, Giulio and Barasuol, Victor. and Semini, Claudio}, title = {SafeSteps: Learning Safer Footstep Planning Policies for Legged Robots via Model-Based Priors}, booktitle = {IEEE-RAS International Conference on Humanoid Robots}, year = {2023}, }