Jordi Salvador

Papers

Total Citations

H-Index

About

Jordi Salvador is a leading researcher in Embodied AI, a field where agents learn to complete tasks by interacting with their environments through egocentric observations. His work bridges deep reinforcement learning, computer vision, and robotics to create more capable and practical embodied agents. Salvador’s most cited paper, "AllenAct: A Framework for Embodied AI Research" (2020, 44 citations), introduced a flexible, modular platform that has become a foundational tool for the community, enabling standardized experimentation and accelerating progress in visual navigation and manipulation tasks. In his more recent work, "Towards Disturbance-Free Visual Mobile Manipulation" (2023), Salvador tackles a critical but often overlooked challenge: building robots that can perform tasks without causing unintended disruptions to their surroundings. By shifting focus from raw speed to safety and minimal environmental impact, this research advances the real-world applicability of robotic systems. Salvador’s contributions are shaping how embodied agents are designed, trained, and evaluated, making him a key figure in the push toward intelligent, disturbance-free robots that can operate safely alongside humans.