Self-awareness and Self-control in NARS

Robotics and Autonomous Systems, 1989

We view an autonomous intelligent being as a total system. We believe that the following main functions are required if the system is to show intelligence: get sensory inputs and describe the present situation, select a sub-objective, construct a plan based on experience, do the actions indicated by the plan. receive the resulting pleasure level from the body and learn from experience. A formal definition of the terms used is given, including "autonomous intelligent system". A program was written, implementing the basic functions of this system. Our experience in running this program is, that after a learning period of about 3000 instants of life in our problem space, it reaches the sub-objectives it has selected for itself. It is noteworthy that an autonomous intelligent system with these characteristics is not limited to a particular problem space, but can reach its objective in any problem space.

2010

To arrive at truly intelligent agents such as robots, many issues need to be addressed such as perception, locomotion, manipulation, human-robot interaction, planning and reasoning. In this thesis we are concerned with the latter in the sense of cognitive robotics as introduced by the late Ray Reiter, meaning “the study of the knowledge representation and reasoning problems faced by an autonomous robot (or agent) in a dynamic and incompletely known world”(Levesque and Lakemeyer (2008)).

Computer, 2015

A dvanced computing systems generally contain many heterogeneous subsystems, each with a local perspective and goal set, which interconnect in changing network topologies. The subsystems must interact with each other and with humans in ways that are difficult to understand and predict while robustly maintaining performance, reliability, and security even with unforeseen dynamics, such as system failures or changing goals. To meet these stringent requirements, computational systemsranging from robot swarms and personal music devices to Web services and sensor networks-must achieve sophisticated autonomous behavior by adapting themselves at runtime and through learning processes that enable ongoing self-change. Managing tradeoffs among conflicting local and global goals at runtime requires considerable awareness of both the system's current state and its environment. Yet researchers have only recently begun to understand the implications of selfawareness principles and how to translate them into system engineering. Consequently, there is no general methodology for architecting self-aware systems or for comparing their self-awareness capabilities. To address this need, we examined how human selfawareness can serve as a source of inspiration for a new notion of computational self-awareness and associated self-expression, and we developed a general framework for describing a computing system's self-awareness properties. As part of this work, we created a reference architecture, which we used to derive architectural patterns for

Self-Aware Computing Systems

Lecture notes in networks and systems, 2022

Abstract The purpose—explain identify the features of General Artificial Intel-ligence (AGI), from the standpoint of scientific rationality stages (classic, non-classical, post-non-classical), and show its difference from traditional Artificial Intel-ligence (AI). The latter is currently implemented mainly with digital computers and implements the functions of recognition, forecasting and preparation of answers to simple questions. New conditions force us to consider AGI from the stand-point of functionalism, as a man–machine system, purposefully functioning in a self-developing poly-subject (reflexive-active) environment. Design/Methodology/Approach—AI development paradigm should consider aspects of its immersion in the social and humanitarian environment and the inno-vative atmosphere. The new paradigm of AI development should reflect the unfor-malized cognitive dynamics of AI models and support the self-development of AI systems under pressure from the external environment. The methodology of creating AGI is based on the ideas of the subject-oriented and ontological approach, func-tionalism, the phenomenology of subjective reality, the convergent cognitive archi-tectures, as well as the methods of creating a self-developing poly-subject (reflexive-active) environment. AGI becomes a hybrid, purposefully integrating the capabilities of a machine and a person. Findings—General properties, value-semantic and intentional-volitional oper-ational structures of the phenomenon of subjective reality do not lean itself on direct formalized and algorithmic representation in discrete computer systems of von Neumann architecture. The study of consciousness in the context of subjective reality made it possible to formulate the main systemic, structural, functional, and operational characteristics of human cognitive activity, which allows a new approach to the modelling of cognitive architectures that meet the tasks of building AGI. The characteristics of subjective reality cannot be fully represented in the paradigm of physicalism; that is, it cannot be represented only with the help of physical devices. The chapter proposes a non-reductionist way of taking this characteristic into account by considering the problem of consciousness in an ontological and epistemological context, which allows representing the processes of consciousness and cognitive activity of a person and a group of people indirectly and inversely. Originality/Value—State-of-the-art cognitive architectures and traditional AI approaches practically ignore solving the problems of AGI. They are more focused on the formalized construction of a thinking model, identifying physical blocks and processes of mental activity. At the same time, for AGI, ontological, subjec-tive and hybrid reality issues are of the most importance, especially in explanations of the activity of consciousness, unconsciousness, and causeless processes, which can act purposefully in conditions of goal uncertainty. AGI must help to describe the phenomena of subjective reality, which causes physical changes, explain the ability of goal setting, free will, the ability of self-management by the physical actions of an individual in a team, etc. Research/Practical/Social/Environment implications—the chapter give rise to a new type of control, which differs from the traditional control in digital reality. The chapter’s results made it possible to uniquely find the optimal measure of centraliza-tion and autonomy of control loops that can ensure the preservation and strengthening of the integrity of a complex poly-subject system functioning in a reflexive-active environment, the interpretation of which does not fit into the narrow framework of digital and algorithmic reality, and traditional AI. Research limitations—the AGI approach based on exceptional methods of constructing subjective reality also has its limitations. For example, the approach we propose to explain the connection between the human brain, consciousness, thought processes and environment does not yet allow us to explain the information and cognitive processes generated by the effect of subjective reality nonlocality, which arises, e.g., at the atomic level of the human brain and should be considered when studying cognitive processes.

International Journal of Machine Consciousness, 2010

Engineering self-conscious robotic systems is a challenge issue because of the intrinsic complexity of such systems; a self-conscious robot has to acquire knowledge, to understand its world and to autonomously interact with its environment. In this paper the externalist point of view is used for developing a complete process for the design and implementation of a conscious robotic system that is able to interact with a dynamic environment in a human like fashion without possessing detailed knowledge about the environment and pre-programmed tasks and algorithms. The paper mainly focusses on the configuration part of the whole process that make the robot able to decide and to learn from experiences.

AI Magazine, 2008

Humans learn from their mistakes. When things go badly, we notice that something is amiss, figure out what went wrong and why, and attempt to repair the problem. Artificial systems depend on their human designers to program in responses to every eventuality and therefore typically don't even notice when things go wrong, following their programming over the proverbial, and in some cases literal, cliff. This article describes our past and current work on the Meta-Cognitive Loop, a domain-general approach to giving artificial systems ...

Technical systems are becoming more complex, they incorporate more advanced func- tionalities, they are more integrated with other systems and they are deployed in less controlled environments. All this supposes a more demanding and uncertain scenario for control systems, which are also required to be more autonomous and dependable. Autonomous adaptivity is a current challenge for extant control technologies. The ASys research project proposes to address it by moving the responsibility for adaptiv- ity from the engineers at design time to the system at run-time. This thesis has intended to advance in the formulation and technical reification of ASys principles of model-based self-cognition and having systems self-handle at run-time for robust autonomy. For that it has focused on the biologically inspired capability of self-awareness, and explored the possibilities to embed it into the very architecture of control systems. Besides self-awareness, other themes related to the envisioned solution have been explored: functional modeling, software modeling, patterns technology, components technology, fault tolerance. The state of the art in fields relevant for the issues of self- awareness and adaptivity has been analysed: cognitive architectures, fault-tolerant control, and software architectural reflection and autonomic computing. The extant and evolving ASys Theoretical Framework for cognitive autonomous systems has been adapted to provide a basement for this selfhood-centred analysis and to con- ceptually support the subsequent development of our solution. The thesis proposes a general design solution for building self-aware autonomous systems. Its central idea is the integration of a metacontroller in the control archi- tecture of the autonomous system, capable of perceiving the functional state of the control system and reconfiguring it if necessary at run-time. This metacontrol solution has been formalised into four design patterns: i) the Metacontrol Pattern, which defines the integration of a metacontrol subsystem, con- trolling the domain control system through an interface provided by its implemen- tation component platform, ii) the Epistemic Control Loop pattern, which defines a model-based cognitive control loop that can be applied to the design of such a meta- controller, iii) the Deep Model Reflection pattern proposes a solution to produce the online executable model used by the metacontroller by model-to-model transforma- tion from the engineering model, and, finally, iv) the Functional Metacontrol pattern, v which proposes to structure the metacontroller in two loops, one for controlling the configuration of components of the controller, and another one on top of the former, controlling the functions being realised by that configuration; this way the functional and structural concerns become decoupled. A reference architecture and an associated metamodel are the core pieces of the architectural framework developed to reify this patterned solution. The metamodel has been developed for representing the structure and its relation to the functional requirements of any autonomous system. The architecture is a blueprint for building a metacontroller according to the patterns. This metacontroller can be integrated on top of any component-based control architecture. At the core of its operation lies a model of the control system that conforms to the metamodel. An engineering process and accompanying assets have been constructed to com- plete and exploit the architectural framework. The engineering process defines the methodology to follow to develop the metacontrol subsystem from the functional model of the controller of the autonomous system. The assets include software li- braries that provide a domain and application-independent implementation of the meta- controller. They can be used in the implementation phase of the methodology. Finally, the complete solution has been validated in the development of an au- tonomous mobile robot that incorporates an metacontroller. The functional self-awareness and adaptivity properties achieved thanks to the metacontrol system have been vali- dated in different scenarios. In these scenarios the robot was able to overcome failures in the control system thanks to reconfigurations performed by the metacontroller.

IEEE/CAA Journal of Automatica Sinica, 2021

Autonomous systems are an emerging AI technology functioning without human intervention underpinned by the latest advances in intelligence, cognition, computer, and systems sciences. This paper explores the intelligent and mathematical foundations of autonomous systems. It focuses on structural and behavioral properties that constitute the intelligent power of autonomous systems. It explains how system intelligence aggregates from reflexive, imperative, adaptive intelligence to autonomous and cognitive intelligence. A hierarchical intelligence model (HIM) is introduced to elaborate the evolution of human and system intelligence as an inductive process. The properties of system autonomy are formally analyzed towards a wide range of applications in computational intelligence and systems engineering. Emerging paradigms of autonomous systems including brain-inspired systems, cognitive robots, and autonomous knowledge learning systems are described. Advances in autonomous systems will pave a way towards highly intelligent machines for augmenting human capabilities.

Cognitive Computation, 2012

This work discusses the challenge of developing self-cognisant artificial intelligence systems, looking at the possible benefits and the main issues in this quest. It is argued that the degree of complexity, variation and specialisation of technological artefacts used nowadays, along with their sheer number, represent an issue that can and should be addressed through an important step towards greater autonomy, that is, the integration of learning, which will allow the artefact to observe its own functionality and build a model of itself. This model can be used to adjust the expectations from an imperfectly manufactured item, patch up its performance, and control its consistency over time, so providing a form of selfcertification and a warning mechanism in case of deterioration. It is suggested that these goals cannot be fully achieved without the ability of the learner to model its own performance, and the implications and issues of this self-reflective learning are debated. A possible way of quantifying the faculty for self-cognition is proposed, and relevant areas of computer science, philosophy, and the study of the evolution of language are mentioned.