Max Robotics — Global Service Robot Directory + US Certification

About

Tara Sadjadpour is a robotics researcher whose work bridges perception, manipulation, and real-world assembly tasks. Her primary research areas include 3D multi-object tracking, interactive perception for robot manipulation, and deformable object assembly. Her most influential contribution, "ShaSTA: Modeling Shape and Spatio-Temporal Affinities for 3D Multi-Object Tracking" (2023), has garnered 29 citations and addresses a critical challenge in autonomous driving: improving tracking quality by modeling object shape and spatio-temporal relationships to handle over-detection safely. This work is foundational for robust perception in safety-critical systems. Sadjadpour also introduced "MANIP: A Modular Architecture for Integrating Interactive Perception for Robot Manipulation" (2024), a systems-level framework that systematically combines learned subpolicies with classical robotics primitives like Inverse Kinematics and Kalman Filters, enabling more reliable and adaptable manipulation. Her applied work, "Automating Deformable Gasket Assembly" (2024), tackles a high-precision, long-horizon manufacturing task common in automotive and electronics industries, demonstrating her ability to solve complex industrial problems. Through these contributions, Sadjadpour is shaping how robots perceive, track, and physically interact with their environments, with clear impact on both autonomous driving and automated manufacturing.