Built to Learn: An Intrinsically Motivated Approach to Incremental Learning in Robotics

• Letícia M. Berto UNICAMP
• Ricardo R. Gudwin UNICAMP
• Esther L. Colombini UNICAMP

DOI: https://doi.org/10.5753/sbrlars_estendido.2025.248224

Resumo

This research explores how robots can autonomously adapt to complex environments using a human-inspired cognitive development framework. By integrating intrinsic motivation and decision-making, robots employed reinforcement learning to balance internal states. An extended model introduced the concept of pleasure, enabling context-aware prioritization of needs and preferences. A child-inspired cognitive architecture guided learning, interaction, and survival, giving rise to distinct robot personalities that shaped human engagement. Building on this, an advanced model with Theory of Mind allowed two robots to collaborate through mutual understanding of needs. We validated our models on both simulated and real robots with varied architectures.

Palavras-chave: Cognitive architecture, human-robot interaction, robotics, intrinsic motivation, theory of mind, reinforcement learning

Referências

Baima, R. L., Berto, L. M., and Roth, T. (2024). Expanding the scope—cognitive robotics meets neurois. In Davis, F. D., Riedl, R., Brocke, J. v., Léger, P.-M., Randolph, A. B., and Müller-Putz, G. R., editors, Information Systems and Neuroscience, pages 195–203, Cham. Springer Nature Switzerland.

Begazo, M. F. T. (2020). A learning-based model-free controller for decoupled humanoid robot walking. Instituto de Ciência e Tecnologia de Sorocaba - Universidade Estadual Paulista ”Júlio de Mesquita Filho”(UNESP) (Master thesis).

Berto, L., Costa, P., Simões, A., Gudwin, R., and Colombini, E. (2024a). A motivational-based learning model for mobile robots. Cognitive Systems Research, 88:101278.

Berto, L., Hellou, M., Sciutti, A., Gudwin, R., Colombini, E., and Cangelosi, A. (2025a). A theory of mind-driven motivational framework for social interaction with autonomous cognitive robots. In 2025 IEEE International Conference on Robot and Human Interactive Communication (RO-MAN). Accepted.

Berto, L., Rossi, L., Costa, P., Simões, A., Gudwin, R., and Colombini, E. (2025b). Where are my toys going? exploring incremental learning in object tracking and permanence. IEEE Transactions on Cognitive and Developmental Systems. Paper in progress.

Berto, L., Rossi, L., Rohmer, E., Costa, P., Gudwin, R., Simões, A., and Colombini, E. (2024b). Piagetian experiments to devrobotics. Cognitive Systems Research, 83:101170.

Berto, L., Tanevska, A., Cirne, A., Costa, P., Simões, A., Gudwin, R., Rea, F., Colombini, E., and Sciutti, A. (2024c). Curiosity and affect-driven cognitive architecture for hri. In 2024 IEEE International Conference on Development and Learning (ICDL). Late-Breaking Results.

Berto, L., Tanevska, A., Cirne, A., Costa, P., Simões, A., Gudwin, R., Rea, F., Colombini, E., and Sciutti, A. (2025c). Curiosity and affect-driven cognitive architecture for hri. IEEE Transactions on Affective Computing, pages 1–18.

Berto, L. M. and Colombini, E. L. (2025a). Human-robot interaction: motivated autonomous robots. DOI: 10.25824/redu/ZWJPCW, Repositório de Dados de Pesquisa da Unicamp, V1.

Berto, L. M. and Colombini, E. L. (2025b). Theory of mind: motivational framework for social interaction with robots. DOI: 10.25824/redu/MVWJGS, Repositório de Dados de Pesquisa da Unicamp, V1.

Berto, L. M. and Colombini, E. L. (2025c). Wanting vs. liking: motivational framework for mobile robots. DOI: 10.25824/redu/BWRUOL, Repositório de Dados de Pesquisa da Unicamp, V1.

Berto, L. M., Costa, P. D. P., Simoes, A. S., Gudwin, R. R., and Colombini, E. L. (2021). An iowa gambling task-based experiment applied to robots: A study on long-term decision making. In 2021 IEEE International Conference on Development and Learning (ICDL), pages 1–6.

Cangelosi, A. and Schlesinger, M. (2015). Developmental Robotics: From Babies to Robots. The MIT Press.

Cannon, W. B. (1939). The wisdom of the body.

G. S. VIRK, S. M. and GELIN, R. (2008). Iso standards for service robots. In Marques, L., de Almeida, A., Tokhi, M. O., and Virk, G. S., editors, Advances in Mobile Robotics, pages 133–138. World Scientific.

Hull, C. L. (1943). Principles of behavior: An introduction to behavior theory.

Lungarella, M., Metta, G., Pfeifer, R., and Sandini, G. (2003). Developmental robotics: a survey. Connection Science, 15(4):151–190.

Maslow, A. H. (1943). A theory of human motivation. Psychological review, 50(4):370.

Maslow, A. H. (1981). Motivation and personality. Prabhat Prakashan, New Delhi.

Piaget, J. (1952). The origins of intelligence in children. The origins of intelligence in children. W W Norton & Co, New York, NY, US.

Rossi, L., Berto, L., Costa, P., Gudwin, R., Colombini, E., and Simões, A. (2025). Dual or unified: Optimizing drive-based reinforcement learning for cognitive autonomous robots. Cognitive Systems Research. First round review.

Rossi, L., Berto, L., Rohmer, E., Costa, P., Gudwin, R., Colombini, E., and Simões, A. (2022). Aprendizado procedimental e sensório-motor em robôs cognitivos. In Anais Estendidos do XIV Simpósio Brasileiro de Robótica e XIX Simpósio Latino-Americano de Robótica, pages 73–84, Porto Alegre, RS, Brasil. SBC.

Rossi, L. L., Berto, L., Costa, P. D. P., Gudwin, R., Colombini, E., and Simões, A. (2024). Drives and impulses: Shaping motivation and procedural learning for humanoid robots. In 2024 IEEE International Conference on Development and Learning (ICDL), pages 1–8.

Turing, A. M. (1950). Computing machinery and intelligence. Mind, 59(236):433–460.

Vernon, D., von Hofsten, C., and Fadiga, L. (2016). Desiderata for developmental cognitive architectures. Biologically Inspired Cognitive Architectures, 18:116–127.

Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes, volume 86. Harvard university press.

Áureo Marques, Coletta, L., Silva, A., Paraense, A., Berto, L., Costa, P., Colombini, E., Simões, A., and Gudwin, R. (2022). Visualization tools for monitoring and debugging a cognitive architecture using cst. Procedia Computer Science, 213:528–535. 2022 Annual International Conference on Brain-Inspired Cognitive Architectures for Artificial Intelligence: The 13th Annual Meeting of the BICA Society.