Max Robotics — Global Service Robot Directory + US Certification

About

Mark Tjersland is a roboticist focused on enabling robust manipulation in unstructured, real-world environments—from cluttered homes to busy grocery stores. His work centers on perception, motion planning, and system integration for mobile manipulation. A key contribution is **SimNet**, a stereo-based perception system trained purely on synthetic data that achieves robust manipulation of challenging unknown objects, including transparent and reflective items (14 citations). He also developed a learned stereo depth system optimized for human environments, producing dense, high-resolution point clouds even on dark, textureless, or specular surfaces. Tjersland’s impact extends to full-system validation: he demonstrated a general-purpose mobile manipulation platform in an unmodified grocery store, benchmarking performance through a metrics-driven approach (13 citations). To achieve reliable, sub-second motion planning in changing environments, he leveraged large-scale dynamic roadmaps for high-degree-of-freedom robots. His work on autonomous vehicle configuration management also addresses critical challenges in safety and repeatability. Tjersland’s research bridges the gap between simulation and real-world deployment, advancing the practical capabilities of robots to operate robustly in human-centric spaces.