Dear All,
We are pleased to invite you to the IEOR Seminar by Rahul Mangharam on 11 August 2025.
Date and time: 11 August 2025 (Monday), 10:30 – 11:30 a.m.
Venue: IEOR Seminar Room
Speaker: Rahul Mangharam, University of Pennsylvania
Title: RetroLift: Low-Cost Infrastructure-Assisted Autonomy for Forklifts
Abstract:
Traditional Autonomous Mobile Robots (AMRs) for material handling are cost-prohibitive, unreliable in dynamic warehouse environments, and ill-suited for third-party logistics operations. In this session, we introduce RetroLift, an infrastructure-based autonomy system that retrofits existing drive-by-wire forklifts with low-cost sensors and minimal onboard compute. RetroLift shifts the burden of perception and planning to static facility-mounted cameras and compute nodes, leveraging a hybrid of Birds-Eye-View (BEV) and First-Person-View (FPV) to enable fleet-scale autonomous teleoperation. We detail our approach, which reduces deployment costs by 2–3x, avoids the brittleness of AMRs in cluttered crossdocks, and allows seamless integration with existing manual operations. The talk will cover simulation studies, initial prototype deployments, and the roadmap toward scalable, 24/7 autonomous forklift fleets. RetroLift offers a pragmatic path to end-to-end warehouse automation with flexibility, safety, and economic feasibility.
Bio:
Rahul Mangharam is an Associate Professor at the University of Pennsylvania, where he leads cutting-edge research at the intersection of robotics, cyber-physical systems, and machine learning. As the founder of Jugaad Labs LLC, he applies his research toward solving practical, large-scale automation challenges in logistics, mobility, and industrial operations. His work on autonomous racing and multi-agent systems under uncertainty has received global recognition, and he has led numerous international competitions to crowdsource robust control algorithms. Through initiatives like RetroLift, Rahul is pioneering infrastructure-based autonomy solutions that lower the barriers to intelligent automation in brownfield industrial environments.