Max Robotics — Global Service Robot Directory + US Certification

About

Assem Sadek is a robotics researcher focused on bridging classical and learning-based approaches for visual navigation. His work centers on enabling mobile robots to navigate real-world environments by integrating Simultaneous Localization and Mapping (SLAM) with deep reinforcement learning and imitation learning. Sadek’s key contributions include developing hybrid policies that dynamically switch between classical planning and neural networks, allowing robots to leverage the reliability of traditional methods while benefiting from the adaptability of machine learning. His 2022 study on sensor usage and visual reasoning (8 citations) provides foundational insights into how robots can combine perceptual data with high-level reasoning for robust navigation in physical spaces. In subsequent work, Sadek introduced transferable latent spatial representations that enable navigation without explicit map reconstruction, and explored trust-based switching mechanisms between planning paradigms. His research on multi-object navigation in real environments (6 citations) demonstrates practical applications of these hybrid approaches. By systematically addressing the limitations of purely classical or purely learned navigation, Sadek is advancing the development of more reliable and intelligent autonomous systems capable of operating in complex, unstructured settings.