VDML is an independent research lab in San Jose, California, studying machine autonomy: why and how a machine can discover, acquire, and compose new skills on its own, without human intervention.

Unlike research programs that focus on replicating human cognition and integrating autonomy into automation, this lab examines autonomy from the machine's own perspective. Granting autonomy to machines has profound impacts on our society, yet the principles and consequences are not well understood. By uncovering the fundamental systems and processes underlying the emergence of autonomous behavior in non-human agents, VDML aims to help lay the foundations of a more transparent future with machine autonomy.

Principal Investigator: Satoru Isaka, Ph.D. (UCSD, Systems Science)

ORCID Scholar ResearchGate IEEE Xplore GitHub LinkedIn

Research

What does it take for a machine to learn a new skill on its own?

Our work centers on the conditions under which a non-human agent can discover, acquire, and compose new behaviors without human demonstration or task-specific programming. We posit that value systems, innately given or circumstantially shaped, supply the purpose of autonomy, and in turn drive the successive development of dynamic memory functions. This results in progressive changes in the behavior of a physical sensorimotor system, from innate reflexive to purposeful behavior, and eventually conceptual and social behavior.

How should machine autonomy be architected?

A core question we pursue is: what does it mean to grant autonomy to machines? We examine this question through four essential queries: what can machines know? (question of knowledge), what can machines do? (question of behavior), what constitutes autonomy? (question of identity), and what does it mean for machines to be autonomous? (question of freedom). Working through these queries, we distill the core principles and build the structural and representational frameworks of machine autonomy.

What does machine autonomy look like in practice?

We build and evaluate concrete systems, in simulation and on hardware, that demonstrate end-to-end autonomous skill acquisition and developmental behavior. We specifically focus on the nature of temporal, in-the-moment experience in machines. Unlike computational models for images and language, sound is inherently ephemeral and embodied, making it a natural medium for interpretation, expression, motion, and interaction. It challenges machines to engage in self-directed interpretation and kinematic expression, rather than predetermined execution.

Read the full research outline Internal report · Aug 2025.

Publications

2026
S. Isaka, "The Architecture of Mechanical Intelligence: A Technical Reference for Actionable AI Policy," (under review)
S. Isaka, "Acting on the Ephemeral: Thought Experiments on Ontologically Grounded Behavior in Machines," (in preparation)
2025
S. Isaka, "Taxonomic Robot Identifiers: Toward General Classification and Oversight for Autonomous Systems," in IEEE Access, vol. 13, pp. 101801–101816, doi: 10.1109/ACCESS.2025.3578870
S. Isaka, "Philosophical Foundations of Machine Autonomy: What It Means To Grant Autonomy To Machines — Part 1: Introduction and First Query," Preprint, doi: 10.13140/RG.2.2.10280.48643
S. Isaka, "Critique of Emerging Linguistic Centralism: How Language Distorts Our Understanding of AI and Humanity," Preprint, doi: 10.13140/RG.2.2.27676.96643
S. Isaka, "Research Outline: Machine Autonomy and Developmental Autonomous Behavior," Internal report, August 2025. [PDF] · doi: 10.13140/RG.2.2.20288.24325
2024
S. Isaka, "Autonomy in Cognitive Development of Robots: Embracing Emergent and Predefined Knowledge and Behavior," in Proc. IEEE 20th Int. Conf. on Automation Science and Engineering (CASE), Bari, Italy, pp. 1353–1360, doi: 10.1109/CASE59546.2024.10711540
S. Isaka, "A Taxonomic Classification and Identification System for Robots: Abstract," in Proc. 2024 IEEE Int. Conf. on Systems, Man, and Cybernetics (SMC), Kuching, Malaysia, pp. 3799–3800, doi: 10.1109/SMC54092.2024.10831650
2023
S. Isaka, "Developmental Autonomous Behavior: An Ethological Perspective to Understanding Machines," in IEEE Access, vol. 11, pp. 17375–17423, doi: 10.1109/ACCESS.2023.3246840
S. Isaka, "An Ethological Analysis of Developmental Behavior in Machines," in Proc. 2023 IEEE Int. Conf. on Development and Learning (ICDL), Macau, China, pp. 79–86, doi: 10.1109/ICDL55364.2023.10364472
Foundational Research & Previous Works (1987–2018)

Earlier publications establishing the technical foundations in adaptive control, fuzzy systems, and biomedical engineering that inform current work on machine autonomy.

2018
S. Isaka, The Principles of Mental Care, Independent Publishing.
S. Isaka, Practical English Training with Speech Recognition, Independent Publishing.
2005
S. Isaka and H.T. Nguyen, "Information triage for health literacy promotion," Abstract, 133rd Annual Meeting, American Public Health Association, Philadelphia, PA.
1997
S. Isaka, "An empirical study on facial image feature extraction using genetic programming," Late Breaking Papers, Genetic Programming 1997, Stanford, pp. 93–99.
1995
S. Isaka and V. Chu, "Industrial fuzzy control review: a perspective from feedback and manufacturing," in Industrial Applications of Fuzzy Logic, World Scientific Publishing.
S. Isaka, "Fuzzy logic applications at OMRON," in Industrial Fuzzy Control and Intelligent Systems, IEEE Press.
1993
B. Kosko and S. Isaka, "Fuzzy Logic," Scientific American, vol. 269, no. 1, pp. 76–81.
S. Isaka and A.V. Sebald, "Control strategies for arterial blood pressure regulation," IEEE Trans. Biomedical Engineering, vol. 40, no. 4, pp. 353–363.
S. Isaka, "Fuzzy temperature controller and its applications," Proc. SPIE Applications of Fuzzy Logic Technology, pp. 59–65.
Z.Y. Zhao, M. Tomizuka, and S. Isaka, "Fuzzy gain scheduling of PID controllers," IEEE Trans. Systems, Man, and Cybernetics, vol. 23, no. 5, pp. 1392–1398.
K. Mitsubuchi, S. Isaka, and Z.Y. Zhao, "A fuzzy rule generation," Proc. IFSA World Congress '93, Seoul, pp. 11–14.
1992
S. Isaka, "On neural approximation of fuzzy systems," Proc. IJCNN '92, Baltimore, pp. 263–268.
S. Isaka, "On function approximations by fuzzy and neural systems," Proc. Int'l Symposium on AI in Material Processing, Edmonton (invited).
S. Isaka, "On input space clustering by fuzzy systems and neural networks," Proc. IEEE Int'l Conf. SMC, Chicago (invited).
Z.Y. Zhao, M. Tomizuka, and S. Isaka, "Fuzzy gain scheduling of PID controllers," Proc. IEEE Conf. Control Applications, Dayton, pp. 698–703.
1990
S. Isaka and A.V. Sebald, "An optimization approach for fuzzy controller design," Proc. American Control Conference, pp. 1485–1490.
1989
S. Isaka, "On the design and architecture of adaptive fuzzy controllers and their application to a biomedical control problem," Ph.D. Thesis, UC San Diego.
S. Isaka and A.V. Sebald, "An adaptive fuzzy controller for blood pressure regulation," Proc. IEEE EMBS Conf.
A.V. Sebald, M. Quinn, N.T. Smith, A. Karimi, G. Schnurer, and S. Isaka, "Engineering implications of closed-loop control during cardiac surgery," J. Clinical Monitoring.
A. Karimi, A.V. Sebald, and S. Isaka, "Use of Simulated Annealing in Design of Very High Dimensioned Minimax Adaptive Controllers," Proc. Asilomar Conf. Signals, Systems and Computers.
1988
S. Isaka, A.V. Sebald, A. Karimi, N.T. Smith, and M.L. Quinn, "On the design and performance evaluation of adaptive fuzzy controllers," Proc. IEEE Conf. Decision and Control, pp. 1068–1069.
S. Isaka and A.V. Sebald, "A fuzzy blood pressure controller," Proc. IEEE EMBS Conf., pp. 1410–1411.
1987
S. Isaka, A.M. Schneider, P. Filia, K.K. Lue, R.D. Coutts, and V.L. Nickel, "A new tool for stroke rehabilitation study," Proc. 10th Annual Conf. Rehabilitation Technology, pp. 287–289.

Contact

sisaka@visiondelmar.com