Vanderbilt University University of Missouri-Columbia A Biologically Inspired Adaptive Working Memory for Robots Marjorie Skubic and James M. Keller University.

Slides:



Advertisements
Similar presentations
Distinctive Image Features from Scale-Invariant Keypoints
Advertisements

Pat Langley Computational Learning Laboratory Center for the Study of Language and Information Stanford University, Stanford, California
CogRIC Workshop Adaptive Working Memory: From Computational Neuroscience Model To Robot Control Module David C. Noelle Assistant Professor of Computer.
Cognitive Systems, ICANN panel, Q1 What is machine intelligence, as beyond pattern matching, classification and prediction. What is machine intelligence,
Distinctive Image Features from Scale-Invariant Keypoints David Lowe.
Programming Paradigms and languages
Road-Sign Detection and Recognition Based on Support Vector Machines Saturnino, Sergio et al. Yunjia Man ECG 782 Dr. Brendan.
Presented by Xinyu Chang
Breakout session B questions. Research directions/areas Multi-modal perception cognition and interaction Learning, adaptation and imitation Design and.
Chapter Thirteen Conclusion: Where We Go From Here.
Intelligent Agents Russell and Norvig: 2
Pre-frontal cortex and Executive Function Squire et al Ch 52.
Computational Intelligence Research in ECE Target detection and Recognition Linguistic Scene Description Sketch Understanding What’s Coming? Jim Keller,
MICHAEL MILFORD, DAVID PRASSER, AND GORDON WYETH FOLAMI ALAMUDUN GRADUATE STUDENT COMPUTER SCIENCE & ENGINEERING TEXAS A&M UNIVERSITY RatSLAM on the Edge:
Object Recognition using Invariant Local Features Applications l Mobile robots, driver assistance l Cell phone location or object recognition l Panoramas,
Spatial Reasoning with Guinness References Acknowledgements University of Missouri, Columbia, MO.
Object Recognition with Invariant Features n Definition: Identify objects or scenes and determine their pose and model parameters n Applications l Industrial.
Some concepts from Cognitive Psychology to review: Shadowing Visual Search Cue-target Paradigm Hint: you’ll find these in Chapter 12.
ECE 7340: Building Intelligent Robots QUALITATIVE NAVIGATION FOR MOBILE ROBOTS Tod S. Levitt Daryl T. Lawton Presented by: Aniket Samant.
Spatial Reasoning for Semi-Autonomous Vehicles Using Image and Range Data Marjorie Skubic and James Keller Students: Sam Blisard, George Chronis, Grant.
ADVISE: Advanced Digital Video Information Segmentation Engine
A Study of Approaches for Object Recognition
Object Recognition with Invariant Features n Definition: Identify objects or scenes and determine their pose and model parameters n Applications l Industrial.
EE141 1 Broca’s area Pars opercularis Motor cortexSomatosensory cortex Sensory associative cortex Primary Auditory cortex Wernicke’s area Visual associative.
Tracking multiple independent targets: Evidence for a parallel tracking mechanism Zenon Pylyshyn and Ron Storm presented by Nick Howe.
Object Recognition Using Distinctive Image Feature From Scale-Invariant Key point D. Lowe, IJCV 2004 Presenting – Anat Kaspi.
Marakas: Decision Support Systems, 2nd Edition © 2003, Prentice-Hall Chapter Chapter 7: Expert Systems and Artificial Intelligence Decision Support.
Scale Invariant Feature Transform (SIFT)
How facilitation influences an attractor model of decision making Larissa Albantakis.
Distinctive Image Features from Scale-Invariant Keypoints David G. Lowe – IJCV 2004 Brien Flewelling CPSC 643 Presentation 1.
Biologically Inspired Robotics Group,EPFL Associative memory using coupled non-linear oscillators Semester project Final Presentation Vlad TRIFA.
Scale-Invariant Feature Transform (SIFT) Jinxiang Chai.
Distinctive Image Features from Scale-Invariant Keypoints By David G. Lowe, University of British Columbia Presented by: Tim Havinga, Joël van Neerbos.
A Review of Children, Humanoid Robots and Caregivers (Arsenio, 2004) COM3240 – Week 3 Presented by Gizdem Akdur.
Neural mechanisms of Spatial Learning. Spatial Learning Materials covered in previous lectures Historical development –Tolman and cognitive maps the classic.
An Architecture for Empathic Agents. Abstract Architecture Planning + Coping Deliberated Actions Agent in the World Body Speech Facial expressions Effectors.
Active Vision Key points: Acting to obtain information Eye movements Depth from motion parallax Extracting motion information from a spatio-temporal pattern.
Object Tracking/Recognition using Invariant Local Features Applications l Mobile robots, driver assistance l Cell phone location or object recognition.
 The most intelligent device - “Human Brain”.  The machine that revolutionized the whole world – “computer”.  Inefficiencies of the computer has lead.
Towards Cognitive Robotics Biointelligence Laboratory School of Computer Science and Engineering Seoul National University Christian.
Prediction in Human Presented by: Rezvan Kianifar January 2009.
NEURAL NETWORKS FOR DATA MINING
Report on Intrusion Detection and Data Fusion By Ganesh Godavari.
Background The physiology of the cerebral cortex is organized in hierarchical manner. The prefrontal cortex (PFC) constitutes the highest level of the.
Lecture 2b Readings: Kandell Schwartz et al Ch 27 Wolfe et al Chs 3 and 4.
Robotica Lecture 3. 2 Robot Control Robot control is the mean by which the sensing and action of a robot are coordinated The infinitely many possible.
黃文中 Introduction The Model Results Conclusion 2.
Evolution of Control-Related Mental Models Crystal A. Brandon.
Chapter 50 The Prefrontal Cortex and Executive Brain Functions Copyright © 2014 Elsevier Inc. All rights reserved.
University of Windsor School of Computer Science Topics in Artificial Intelligence Fall 2008 Sept 11, 2008.
DARPA ITO/MARS Project Update Vanderbilt University A Software Architecture and Tools for Autonomous Robots that Learn on Mission K. Kawamura, M. Wilkes,
Comp 15 - Usability and Human Factors
Introduction to Neural Networks and Example Applications in HCI Nick Gentile.
Chapter 5 Multi-Cue 3D Model- Based Object Tracking Geoffrey Taylor Lindsay Kleeman Intelligent Robotics Research Centre (IRRC) Department of Electrical.
Learning Agents MSE 2400 EaLiCaRA Spring 2015 Dr. Tom Way.
Foundations of Information Systems in Business. System ® System  A system is an interrelated set of business procedures used within one business unit.
 A navigational display should serve these four different classes of tasks:  Provide guidance about how to get to a destination  Facilitate planning.
Rational Agency CSMC Introduction to Artificial Intelligence January 8, 2004.
How conscious experience and working memory interact Bernard J. Baars and Stan Franklin Soft Computing Laboratory 김 희 택 TRENDS in Cognitive Sciences vol.
Symbolic Reasoning in Spiking Neurons: A Model of the Cortex/Basal Ganglia/Thalamus Loop Terrence C. Stewart Xuan Choo Chris Eliasmith Centre for Theoretical.
Learning Fast and Slow John E. Laird
Visual Learning with Navigation as an Example
Lecture 07 13/12/2011 Shai Avidan הבהרה: החומר המחייב הוא החומר הנלמד בכיתה ולא זה המופיע / לא מופיע במצגת.
Paper – Stephen Se, David Lowe, Jim Little
CS201 Lecture 02 Computer Vision: Image Formation and Basic Techniques
Architecture Components
Developing an Instructional Strategy
Brief Review of Recognition + Context
Pattern Recognition Binding Edge Detection
Computer Vision in Cell Biology
Presentation transcript:

Vanderbilt University University of Missouri-Columbia A Biologically Inspired Adaptive Working Memory for Robots Marjorie Skubic and James M. Keller University of Missouri-Columbia David Noelle, Mitch Wilkes and Kazuhiko Kawamura Vanderbilt University

University of Missouri-Columbia Outline The role of working memory in cognitive systems Incorporating a human-inspired WM into robots Enabling components for robotic embodiment –Central Executive –Interactive Spatial Language –SIFT Object Recognition –Pre-attentive Vision System Conclusions Demo available

Vanderbilt University University of Missouri-Columbia Working Memory

Vanderbilt University University of Missouri-Columbia Working Memory Working memory systems are those that actively maintain transient information that is critical for successful decision-making in the current context. A working memory system can be viewed as a relatively small cache of task relevant information that is strategically positioned to efficiently influence behavior.

Vanderbilt University University of Missouri-Columbia Robotic Working Memory ● ● The highly limited capacity of working memory, along with its tight coupling with deliberation mechanisms, might alleviate the need for costly memory searches. ● ● Information needed to fluently perform the current task is temporarily kept “handy” in the working memory store. Could robot control systems benefit from the inclusion of a working memory system? Can computational neuroscience models of the working memory mechanisms of the human brain shed light on the design of a robotic working memory system?

Vanderbilt University University of Missouri-Columbia Potential Uses ● ● Focus attention on the most relevant features of the current task. ● ● Guide perceptual processes by limiting the perceptual search space. ● ● Provide a focused short-term memory to prevent the robot from being confused by occlusions. ● ● Provide robust operation in the presence of distracting irrelevant events.

Vanderbilt University University of Missouri-Columbia Adaptive Working Memory ● ● Hand Coding – For relatively routine and well understood tasks, designers may hand code procedures for the identification of useful chunks. ● ● Learning – If the robot is expected to flexibly respond in novel task situations, or even acquire new tasks, it would be beneficial to have a means to learn when to store a particular chunk in working memory. How does the working memory system know when a given chunk of information should be actively maintained in working memory? The central focus of this project is on assessing the utility of adaptive working memory mechanisms for robot control.

Vanderbilt University University of Missouri-Columbia Adaptive Working Memory In The Brain ● ● A number of brain regions are implicated as important components of the human working memory system. ● ● One important region is dorsolateral portions of prefrontal cortex. ● ● Working memory is exhibited in delay period activity. ● ● Cells have been found which encode for locations, visual features, and association rules.

Vanderbilt University University of Missouri-Columbia Recurrence How are high neural firing rates sustained over a delay? Mutual excitation of neurons. Dense recurrent connections in prefrontal cortex. Stripe sets. Attractor network computational models.

Vanderbilt University University of Missouri-Columbia Controlling Updating How does the working memory system know when to actively maintain a given chunk? How does it know when to abandon a previously maintained chunk? The dynamics of recurrent attractor networks are insufficient to meet the simultaneous constraints of (1) active maintenance in the face of distraction and (2) rapid updating when needed. A dynamic gating mechanism is needed.

Vanderbilt University University of Missouri-Columbia The Dopamine System

Vanderbilt University University of Missouri-Columbia Temporal Difference (TD) Learning Change in expected reward is called the temporal difference (TD) error (delta). It is the value that drives learning in a powerful form of reinforcement learning called Temporal Difference (TD) Learning.

Vanderbilt University University of Missouri-Columbia The Actor-Critic Framework (Barto, Sutton, & Anderson, 1983) Actor (policy function) Adaptive Critic (value function) Fixed Critic (reinforcer) Sensory System Motor System External Environment r

Vanderbilt University University of Missouri-Columbia TD & Neural Networks TD(0) may be implemented in a connectionist framework, allowing for large continuous state and action spaces and generalization to novel states. The delta value may be used as the error signal for an adaptive critic network learning to produce and also as the error signal for a competitive actor network which implements the policy. Sensory Inputs Sensory Inputs Actions Critic:Actor:

Vanderbilt University University of Missouri-Columbia Dopamine & Working Memory ● ● The dopamine system may be encoding a TD error signal which is useful for learning sequential behaviors. (Montague, Dayan, & Sejnowski) ● ● If the dopamine system can be used to learn to choose overt actions, why couldn't it be used to choose covert actions, such as deciding when to close the gate on working memory contents? (Braver & Cohen) – – There are extensive dopamine projections to PFC. – – There is some evidence that dopamine may influence PFC neurons in a manner consistent with “gating”.

Vanderbilt University University of Missouri-Columbia The Working Memory Toolkit ● ● Memory traces or chunks will be pointers to arbitrary C++ data structures. ● ● The adaptive working memory toolkit will require the user to specify: – – the capacity of the working memory – – a function which extracts features from chunks – – a function which provides relevant features of the current system state, including candidate chunks – – a function which provides instantaneous external reward information ● ● The toolkit provides a function for examining the contents of working memory, returning chunk pointers.

Vanderbilt University University of Missouri-Columbia Critical Related Technologies Feature extraction is critical for success! Advances in perception systems are needed to extract appropriate high level features from experiences. –Guide attention to relevant aspects of experiences. –Identify features associated with objects or object categories. –Identify important qualitative spatial relationships. Advances in motor control systems are needed to fully leverage the benefits of an adaptive working memory.

Vanderbilt University University of Missouri-Columbia Delayed Saccade Task

Vanderbilt University University of Missouri-Columbia Enabling components for robotic embodiment Central Executive

Vanderbilt University University of Missouri-Columbia A Humanoid Cognitive Robot A cognitive robot has the capacity to reflect and generalize to new situations in a complex, changing world. Toward this goal, we have implemented numerous memory structures within an agent-based system. ISAC

Vanderbilt University University of Missouri-Columbia Central Executive Multiagent-based Cognitive Robot Architecture In this project, we concentrate on the Central Executive (CE) and the Working Memory System (WMS) which are two key elements of Cognitive Control

Vanderbilt University University of Missouri-Columbia Cognitive Control Mechanism for intelligent behavior selection and control Behaviors are selected based on task context and past experience Central Executive (CE) –Selects and loads candidate chunks (behaviors) into the WM –Controls task execution of loaded behaviors –Evaluates and updates criteria for selection and control Working Memory System (WMS) –Maintains task related info –Focuses on execution of current task

Vanderbilt University University of Missouri-Columbia Working Memory   Behaviors are loaded into the WMS based on past experience   A behavior consists of a State Estimator which predicts the next system state, and a Controller which issues actual motor commands. Action Selection   Behaviors are executed based on goal related information Action Selection in a Cognitive Robot

Vanderbilt University University of Missouri-Columbia Initial WM Experiment A set of task-related behaviors is taught to ISAC. For the task, ISAC is asked to reach to a point on the table. ISAC must select correct behaviors and combine them in order to perform the task successfully Later, ISAC will be asked to identify and point to an object on the table Goal Position Blue lines denote loaded candidate behavior motions. Red dotted line denotes final behavior motion

Vanderbilt University University of Missouri-Columbia Enabling components for robotic embodiment Central Executive Interactive Spatial Language

Vanderbilt University University of Missouri-Columbia Interactive Spatial Language Cognitive models indicate that people use spatial relationships in navigation and other spatial reasoning (Previc, Schunn) More natural interaction with robots Spatial language can be used to: – Focus attention “look to the left of the telephone” – Issue commands “pick up the book on top of the desk” – Describe a high level representation of a task “go behind the counter, find my coffee cup on the table, and bring it back to me” – Receive feedback from the robot describing the environment “there is a book on top of the desk to the right of the coffee cup

Vanderbilt University University of Missouri-Columbia Our Spatial Modeling Tool Capturing qualitative spatial information between 2 objects The histogram of constant forces The histogram of gravitational forces Features extracted from the histograms are used to generate linguistic spatial terminology Matsakis et al. 1999, 2001

Vanderbilt University University of Missouri-Columbia The tower is perfectly to the LEFT of the stack buildings. Linguistic Scene Description Example

Vanderbilt University University of Missouri-Columbia The group of storehouses is loosely ABOVE-LEFT of the stack buildings. Linguistic Scene Description Example

Vanderbilt University University of Missouri-Columbia The storehouse is ABOVE the stack buildings, but shifted to the RIGHT. Linguistic Scene Description Example

Vanderbilt University University of Missouri-Columbia HRI Scenario: Scene 1 Human : “How many objects do you see?” Robot :“I am sensing 4 objects.” Human :“Where are they?” Robot :“There are objects in front of me. I am surrounded from the rear. The object # 4 is to the right of me.” Human :“Where is the nearest object in front of you?” Robot :“The object # 3 is in front of me but extends to the right relative to me. The object is very close.”

Vanderbilt University University of Missouri-Columbia HRI Scenario: Scene 1, cont. Human : “Object #3 is a box.” Robot :“I now know object#3 is a box.” Human :“Go to the right of the box.”

Vanderbilt University University of Missouri-Columbia HRI Scenario: Scene 2 The robot has moved to the right of the box. The pillar has now come into view (object #4)

Vanderbilt University University of Missouri-Columbia Spatial Language Validated with User Studies Guinness

Vanderbilt University University of Missouri-Columbia Between two objects

Vanderbilt University University of Missouri-Columbia Extend to 3D by Combining the Horizontal and Vertical Planes Look for the coffee cup on top of the desk to the right of the computer. Continue user studies to validate the algorithms

Vanderbilt University University of Missouri-Columbia Use WM to Find Jim’s Coffee Cup Use the working memory toolkit to test a global spatial representation vs. a relational spatial representation

Vanderbilt University University of Missouri-Columbia Enabling components for robotic embodiment Central Executive Interactive Spatial Language SIFT Object Recognition

Vanderbilt University University of Missouri-Columbia Find features that are invariant to image scaling, translation, and rotation, and partially invariant to illumination changes and affine transformations or 3D projection Create Keypoints from extrema in scale space Generate relative position features (naturally translation invariant) Compute directional histograms that are invariant to rotation –Method of calculation also gives insensitivity to affine stretches Normalization helps with Illumination Changes Scale Invariant Feature Transform (SIFT) for Object Recognition Based on the work by David Lowe

Vanderbilt University University of Missouri-Columbia Gaussian Blurring and Differencing Hunt for local extrema in space and scale Keypoint locations on training image Keypoint Descriptions Major direction of gradients is determined Rotate gradient locations so that keypoint orientation is 0º. Rotate individual gradient directions to be consistent with orientation Directional Histograms Sixteen Gradient Histograms Created

Vanderbilt University University of Missouri-Columbia Recognition Examples Top Images Are Training; Bottom Are Test Still matches Keypoints on occluded objects

Vanderbilt University University of Missouri-Columbia Stereo Vision Left Eye Right Eye Keypoints Matching

Vanderbilt University University of Missouri-Columbia 3D Representation for Spatial Relations The scene 3D keypoints projected onto the horizontal and vertical planes

Vanderbilt University University of Missouri-Columbia Can We Use WM to Learn Interesting Landmarks? Use Keypoint Clusters to Determine Potential Areas of Interest Must eliminate the concentration of keypoints along the skyline

Vanderbilt University University of Missouri-Columbia Enabling components for robotic embodiment Central Executive Interactive Spatial Language SIFT Object Recognition Pre-attentive Vision System

Vanderbilt University University of Missouri-Columbia Pre-attentive Vision System Goals Learn broad categories of objects from experience. Be able to explain how it makes decisions, as well as to justify any particular decision. Detect if there are novel elements in a visual scene, and use this to trigger new learning, i.e., self-directed learning. After making a general class identification, use other object recognition algorithms to identify a specific object.

Vanderbilt University University of Missouri-Columbia Elements of Pre-attentive Vision System Feature vectors consist of a color histogram of 250 colors and a measure of texture roughness, 251 features total Fuzzy rules extracted from training data ML estimator for classes Perceptual memory of past experiences Interaction interface for teaching and assessment

Vanderbilt University University of Missouri-Columbia Novelty Detection Train the system on the empty scene. Add new elements to the scene. Identify the new elements by novelty.

Vanderbilt University University of Missouri-Columbia ML Segmentation Yellow = Sidewalk, Blue = Grass, Red = Tree, Green = Artificial Landmark

Vanderbilt University University of Missouri-Columbia WM Experiment Pre-attentive processing significantly reduces the search space for other algorithms such as SIFT. Use WM to learn the most successful pre- attentive identifications, e.g., which lead to the greatest success in reaching a navigational goal. gravel trees sky

Vanderbilt University University of Missouri-Columbia Conclusions ● Working memory plays an important role in cognitive systems to maintain transient information that is critical for successful decision-making ● A biologically inspired working memory toolkit has been constructed for use on robotic testbeds ● A series of experiments are planned to test the feasibility ● Delayed saccade task ● Learn to select and combine behaviors ● Find Jim’s coffee cup: tests spatial representation ● Learn interesting landmarks for navigation using SIFT keypoints ● Learn successful pre-attentive identifications ● System-level tests incorporating all components

Vanderbilt University University of Missouri-Columbia Acknowledgements Funded by the NSF ITR program (EIA ) Thanks to NRL for the use of –Nautilus: Natural Language Understanding system (Speech recognition by Via Voice) –Mobile robot components for building maps, localization, and path planning Students –MU: Bob Luke, Sam Blisard, Charlie Huggard, Steven Senger –VU: Josh Phillips, Albert Spratley, Palis Ratanaswasd, Will Dodd, Julia High, Mert Tugcu See also: