Max Robotics — Global Service Robot Directory + US Certification

About

Oren Spector is a robotics researcher whose work sits at the intersection of robot learning, contact-rich manipulation, and industrial assembly automation. His research focuses on equipping robots with the dexterity and adaptability needed to perform complex insertion and assembly tasks — challenges that have long limited the broader deployment of industrial robots. Spector is best known for developing the InsertionNet framework, a scalable deep learning approach to robotic insertion that tackles a historically narrow and brittle problem in automation. The original InsertionNet (2021, 49 citations) demonstrated that insertion skills could generalize across diverse shapes and configurations, while InsertionNet 2.0 (2022, 21 citations) advanced this further by enabling rapid, minimal-contact multi-step insertion using multimodal sensory inputs with minimal human intervention. Together, these contributions represent a significant leap toward flexible, real-world assembly automation. Complementing this work, Spector has explored how haptic feedback and compliant motion can be harnessed through reinforcement learning to handle the unpredictable forces inherent in contact-rich tasks, as shown in his research on residual admittance policies and compliant movement primitives (both 2020–2021, 15 citations each). His body of work collectively advances the vision of adaptable, learning-enabled industrial robots capable of handling precision tasks once reserved for skilled human operators.