Joshua Friesen
Papers
1
Total Citations
7
H-Index
1
About
Joshua Friesen is a researcher advancing the frontier of robotic perception and autonomous navigation, with a focus on real-time probabilistic semantic mapping. His most-cited work, "ConvBKI: Real-Time Probabilistic Semantic Mapping Network With Quantifiable Uncertainty" (2024, 7 citations), introduces a modular neural network that achieves real-time (>10 Hz) semantic mapping in uncertain environments. The key innovation lies in explicitly updating per-voxel probabilistic distributions within a neural network layer, bridging the reliability of classical probabilistic algorithms with the efficiency of modern deep learning. This approach provides quantifiable uncertainty estimates—a critical feature for safe decision-making in autonomous systems. Friesen’s contributions address a fundamental challenge in robotics: how to maintain robust environmental understanding under sensor noise and dynamic conditions. By enabling high-speed, uncertainty-aware mapping, his work has direct implications for self-driving cars, drones, and field robots operating in unpredictable settings. His research stands out for its practical emphasis on deployability without sacrificing theoretical rigor, making it a valuable reference for students and engineers working on perception systems. Friesen continues to push the boundaries of how machines interpret and interact with complex, uncertain worlds.
Research Focus
Key Achievements
Top Papers
- 1