Saurav Agarwal
Postdoctoral Researcher
GRASP Laboratory, University of Pennsylvania
Contact: SauravAg@seas.upenn.edu
Postdoctoral Researcher
GRASP Laboratory, University of Pennsylvania
Contact: SauravAg@seas.upenn.edu
Research Focus: Decentralized and Collaborative Intelligent Robotic Systems
Research Interests
My research explores the development of large-scale decentralized and collaborative intelligent systems, enabling robot teams to operate cohesively through advanced environmental perception, inter-robot communication, and coordinated actions. By combining learning-based approaches, robust theoretical frameworks, and practical field implementations, I aim to enhance scalability, efficiency, resilience, and performance for complex robotic tasks. Key technologies in my work include graph neural networks (GNNs), constrained learning, and optimization frameworks.
My Ph.D. research introduced a unified framework for generalized coverage of point, curve, and area features in environments, formalized as graph-based optimization problems. This work led to the design of approximation algorithms with provable guarantees and heuristics for fast, large-scale applications, validated through extensive simulations and real-world experiments. Before my Ph.D., I focused on analyzing and designing mechanisms and parallel robots using optimization techniques. My expertise also spans symbolic algebra systems, open-source library development, project leadership, and student mentorship.
LPAC: Learnable Perception-Action-Communication Loops
A canonical learnable perception-action-communication (LPAC) architecture for decentralized and collaborative intelligent systems. (1) Perception: a CNN processes localized observations and generates an abstract representation. (2) Communication: a graph neural network (GNN) computes what information to share and how to integrate the information received from other robots. (3) Action: a shallow MLP computes control actions based on the GNN output. The three modules are executed on each robot independently, with the GNN facilitating collaboration.
Coverage Control Problem
Collaboratively provide sensor coverage to a convex environment. The robots have limited sensing and communication capabilities.
Robust Generalization and Transfer
Demonstrates strong performance across diverse environments, larger swarms, and noisy robot position estimates. The learned policy scales without retraining or performance degradation.
Real-World Applicability
Validates on a real-world traffic light dataset, showing that the trained LPAC model transfers well to practical scenarios, indicating promise for large-scale, decentralized swarm deployments.
Related Publications:
LPAC: Learnable Perception-Action-Communication Loops with Applications to Coverage Control PDF
S. Agarwal, R. Muthukrishnan, W. Gosrich, V. Kumar, and A. Ribeiro
In review: IEEE Transactions on Robotics (TRO). arXiv:2401.04855 (2024).Asynchronous Perception-Action-Communication with Graph Neural Networks PDF
S. Agarwal, A. Ribeiro, and V. Kumar. arXiv:2309.10164 (2023).Constrained Learning for Decentralized Multi-Objective Coverage Control PDF
J. Cerviño*, S. Agarwal*, V. Kumar, and A. Ribeiro (*equal contribution)
IEEE International Conference on Robotics and Automation (ICRA). arXiv:2409.11311 (2025).Generalizability of Graph Neural Networks for Decentralized Unlabeled Motion Planning PDF
S. Muthusamy, D. Owerko, C. I. Kanatsoulis, S. Agarwal, and A. Ribeiro
arXiv:2409.19829 (2024).
Coverage of Linear Features using Multiple Robots
Wind considerations for energy consumption
The battery life is modeled as a constraint on the length of the tours
Optimize the tours for the total travel cost of all the robots
Amount of data gathered is significantly lesser than current solutions; reduces the computation required to analyze the environment
Applications include inspection of road networks and power lines
Related Publications:
Line Coverage with Multiple Robots: Algorithms and Experiments PDF
S. Agarwal and S. Akella
IEEE Transaction on Robotics (T-RO), vol. 40, pp. 1664-1683, January 2024.The Single Robot Line Coverage Problem: Theory, Algorithms, and Experiments PDF
S. Agarwal and S. Akella
Networks, vol. 82, no. 4, pp. 479–505, June 2023.Approximation Algorithms for the Single Robot Line Coverage Problem PDF bib
S. Agarwal and S. Akella
Algorithmic Foundations of Robotics XIV, Springer International Publishing, Cham, Germany, 2021, pp. 534–550.Line Coverage with Multiple Robots PDF bib
S. Agarwal and S. Akella
IEEE International Conference on Robotics and Automation (ICRA), Paris, France, May 2020.
Variable Formation
Development of algorithms to simultaneously compute the optimal assignments and formation parameters for a team of robots from a given initial formation to a variable goal formation.
The shape of goal formation is provided as input
The scale and location parameters for the goal formation are optimized
Optimal assignments of the robots to the goal positions
The Sum of squared travel distance is minimized
Guaranteed collision-free trajectories
Robots start simultaneously and reach their goal positions simultaneously
O(n^3) running time complexity