Max Robotics — Global Service Robot Directory + US Certification

About

Jaka Sembiring’s research focuses on advancing autonomous mobile robot navigation and localization, with particular expertise in particle filter algorithms, Monte Carlo Localization (MCL), and reinforcement learning. His major contributions center on developing novel resampling mechanisms that enable particle filters to converge faster and become more robust to the “kidnapping problem”—a critical challenge in robotics where a robot is suddenly displaced. His most cited work, “New resampling algorithm for particle filter localization for mobile robot with 3 ultrasonic sonar sensor” (2011, 11 citations), demonstrates this innovation using low-cost ultrasonic sensors. Sembiring also pioneered a simplified Q-learning method to solve Partially Observable Markov Decision Processes (POMDPs) for holonomic mobile robot path planning, reducing value function complexity while maintaining intuitive navigation. His research bridges probabilistic localization and reinforcement learning, offering practical solutions for real-world robotic systems. With a total of 23 citations across his key publications, Sembiring’s work has laid groundwork for more efficient, resilient mobile robot autonomy—particularly valuable for students and researchers exploring sensor-based localization and learning-based navigation in constrained environments.