Jonas Tjomsland

Imperial College London, University of Cambridge

Papers

Total Citations

H-Index

About

Jonas Tjomsland is a robotics and artificial intelligence researcher whose work sits at the intersection of human-robot interaction, social robotics, and machine learning. His research addresses some of the most pressing challenges in making robots genuinely useful and socially aware partners for humans in real-world environments. Tjomsland's most influential contribution, "Mind Your Manners!" (2022, 18 citations), tackled the largely uncharted problem of enabling robots to assess the social appropriateness of their own actions. By introducing a dedicated dataset and framing the challenge as a continual learning problem, his work opened a meaningful new avenue for socially intelligent robotics. His earlier research explored how robots can learn collaborative behaviors directly from human partners using deep reinforcement learning — notably through the sample-efficient Soft Actor-Critic algorithm — demonstrating real-world viability rather than relying solely on simulation. Across his body of work, Tjomsland consistently emphasizes implicit human-robot communication, motor adaptation, and lifelong learning, reflecting a commitment to robots that can evolve alongside human collaborators. With growing citation counts and a research focus highly relevant to both industry and academia, his contributions offer a strong foundation for the next generation of socially competent, collaborative robotic systems.