The Aerial Robotic Transformer (a Campus Challenge Problem Solving Competition organized by NASA) will require embodied intelligence responsible for interaction with environment in an autonomous way, with sensory-motor coordination of its subsystems related to specific mode of operation and mission requirements. It will coordinate sensing, performing active data collection, intelligent data analysis and transmission. It will make autonomous goal oriented decisions when not directed or when facing unpredicted situation, will chose the best mode of operation, and will direct the vehicle to act accordingly. The embodied intelligence system will perform typical pattern recognition tasks like voice and face recognition, path finding, arm-eye coordination, flight and motion control, using sensory inputs and navigational data. In addition, its bio-inspired organization will be capable of learning and goal creation, providing it with self defense and basic capabilities to survive in hostile environment.
In our approach goals are created based on external signals and internally generated stimuli that prompt the machine to do a desired action. A positive outcome of these desired actions will be the reason to learn internal representations, skills and useful associations. The goals emerge gradually from the machine operations. By using its goal creation system the machine defines higher level goals and determines the ways to implement them. The goal creation system generalizes the information received, introducing some degree of invariance for its higher level goals which increases with the level of generalization. Goals are built using a self-organizing hierarchy of spatio-temporal patterns that are the result of deliberate actions and learning.
Anticipation-Based Sequences Learning in Spatio-Temporal Memories
Temporal sequence learning is one of the most critical components for human intelligence. Prediction is an essential element of the temporal sequence learning model. By correct prediction, the machine indicates it knows the current sequence and does not require additional learning. When the prediction is incorrect, learning is executed and the machine learns the new input sequence as soon as the sequence is completed.
In action planning a temporal sequence of planned actions is generated in search of satisfactory solutions (that lowers the pain). At the end of sequence (when solution is discovered) there is a need to remember the steps of this sequence. In the proposed organization, the same network will be responsible for storing and playing back various sequences thus providing a short term memory (STM). While the basic need for this STM is to immediately recall a temporal sequence, it is possible to use this sequence as a training input to the long term memory (LTM). By repeating the same STM sequence for a number of times, LTM can be trained, thus avoiding a requirement of one-shot learning in the LTM.
Models and Methods of Adaptive Control in Intelligent Systems
Intelligent agents develop their understanding and skills through interaction with environment. They explore environment in search of solutions to their goals. While some goals are well defined and easy to measure, others are very complex and require complex evaluation process. It is important to understand how these higher level goals are formulated and what their role in self-organization of memory is. It is also important to understand how multiple goals compete for attention and how they are internally managed. Recently conceived low level model proposes that emergence of higher order level goals is correlated to the emergence of abstract perceptions and complex skills. It is important to develop and test models in which goal creation is an integral part of learning.
It is not clear how multiple goal agents might be organized in relation to each other, how they might come about and self-organize, or how environmental manipulations might facilitate or impede their establishment and organization. We must first establish what models are possible and then use the models to hypothesize contextual (e.g., instruction) manipulations that should facilitate adaptation to and performance in these dynamic contexts. Structural models of embodied intelligence which link goal creation mechanism to perception and action need to be developed and tested. We need to learn how multiple goals can be evaluated and implemented in such structures and how importance and urgency of a goal affects attention shift and selection of an action.
Building Invariant Sensory Representations through Active Vision
The objective of this work is to investigate natural ways of sensory representation building in intelligent systems, by using self-organized learning, integrating continuous observation and saccade movements. The aim for this biologically motivated approach is to achieve visual perception through a retina like sampling of high resolution images with lower resolution artificial retina, and sensory- motor coordination.
The system will use an artificial retina model built up by modeling the rods and cones distributions in human retina. By repeating saccade movements and building their invariant temporal correlations at sufficient detail level, an object will be represented in internal structures. The neural network will use hierarchical feedback structures to build object representations, self-organize invariant transformations, act on the images received from the retina model and control the retina model to sample details of the observed objects. The network will identify the input image by using winner-take-all scheme after sufficiently accurate saccades. By using a unique invariance building scheme, the network will identify different views of the same object. In addition, it will also learn temporal sequences and make predictions.
Hardware Needs for Machine Intelligence
This project focuses on design of self-organizing learning hardware modules for support of studying machine intelligence and spatio-temporal associative learning memory development in cortical minicolumn structures. Such structures will benefit from implementation on regular self-organizing arrays of identical processors with programmable sparse interconnections to other processors and asynchronous data driven operation. An FPGA based system that aims at developing such regular hardware architectures based on modular, expandable 3D architecture is described at the following page:
Self Organizing Learning Array (SOLAR)
- a data driven self-organizing learning hardware for studying machine intelligence, developing its computational models and its structures.
SOLAR is a regular, two or three-dimensional array of identical processing cells, connected to programmable routing channels. Each cell in the array has ability to self-organize by adapting its functionality in response to information contained in its input signals. Cells choose their input signals from the adjacent routing channels and send their output signals to the routing channels.
A SOLAR structure in many ways resembles the organization of cellular neural networks (CNN). Like in a CNN its architecture is defined by an array of identical cells, which adapt their behavior to the input data. Its neurons are cellular automata, which can be programmed to perform different computational tasks based on data received from its neighbors. Neurons can be either static or dynamic, depending on the their implementation and types of signals processed. However, unlike in a CNN, its connectivity structure is not fixed. In a CNN, the interconnect structure is defined by templates which limits its learning ability, while in a SOLAR the interconnect structure is an element of learning and can by dynamically changed even during the network’s operation. Thus a CNN can be considered as a special case of SOLAR structure.
SOLAR has three advantageous features over a typical neural network technology: online learning; dynamically set local interconnections; dynamically set neuron functions and threshold values. Comparing to cellular neural networks, SOLAR has not only dynamically adapting neurons, but dynamically adapting interconnection structure as well.
SOLAR has a hierarchical structure in which data is represented through network topology and neuron's function. It learns in interaction with environment through its interfaces, and stores the useful knowledge in its distributed, hierarchically organized memory. The interfaces include sensory inputs and motor outputs, as well as inputs from reinforcement learning signals. SOLAR is capable of learning through association, and uses associative feedback to predict and screen the incoming information for selective learning of new features.
SOLAR can be used as autonomous control system that uses reinforcement learning and other sensory inputs as the feedbacks from environment in response to its actions. This implementation of SOLAR is meant to be used to study selected aspects of intelligence in interaction with environment through planning and motor functions. There is a close resemblance between the machine anticipation of the result of its action and its motor control. Thus manipulation of environment directed to optimize the state of the machine with respect to its learned value system, together with planning related to finding the anticipated optimum response of the environment is a simple manifestation of intelligent behavior. Learning through interaction with environment builds up machine experience and modifies its value system for better planning and future performance.
At present stage of our research, this learning array can perform intelligent tasks such as pattern recognition, prediction and modeling of unknown systems. It can also learn associations between different input patterns and between different sensors. Its associative learning yields a hierarchical organization of neurons, such that neurons that are farther away from the sensory inputs represent more abstract features or concepts. SOLAR should find a wide range of applications from security, robotics, decision support, information gathering and learning, through everyday life applications in caretaking, monitoring, protection, guidance, to broad applications in military and commercial areas.
Presentations:
Collaborative
Sensing August 2005
Intentional
Robot
SOLAR
simulation
