Copyright © Antonio D'Angelo, Univ. of Udine

Robotica Lectures


(Tentative) Class Schedule






5 Ottobre

Obiectives; What is a Robot; Examples; Sensors, Effectors, Actuators, Controllers; State Type and State Space; Representations Issues; Autonomy and Control Architectures

Introduction to Robotics

6 Ottobre

Distinguishing Issues; Reactive Control; Embodied Intelligence; Cybernetic; Walter's Turtle; Behaviors; Control Paradigms

From Cybernetics to Robotics


12 Ottobre

Robot Control Architectures: deliberative, reactive, behavior-based and hybrid; Feedback and Behavior control, PID controller; Properties; Wall-follow example

Robot Control

Components; Control; Feedback; Open Loop Control; Specification and Performance Parameters; Gain; Proportional, Derivative and Integral Controls; PID Control Issues; Non Linear Feedback, Model-based, Adaptive and Model Reference Adaptive, Sliding Mode and Linear Quadratic Gaussian Controls; H-Infinity Control; Conventional Control Methods Evaluation

Low Level Control

13 Ottobre

Key aspects: situatedness, embodiment, emergence;Issues: grounding, ecological dynamics, scalability; Animal Behaviors; Neuroscience; Psycology: Schema, Behaviorism, Gestalt, cognition; Ethology: reflexes, taxes, fixed-action patterns; behavior/action tradeoff; reactivity

Behavior-based Robotics


19 Ottobre

Key Questions: types, interaction, perception; Behavior template; Innate Realising Mechanism: biological examples; Design Paradigms: ethologically guided, situated activity, experimental design; Combining Behaviors: competitive and cooperative methods; Behavior Notations: response, stimulus domain, mapping; Behavior Null, Discrete, Continuous

Behavior Design

20 Ottobre

Common Features; Evaluation Issues; Subsumption: the world is its own best model; Design: the navigation problem; Arbitration; Examples: Allen; Genghis, Herbert; Subsumption Evaluation

Robot Architectures: Subsumption

Vertical v. Horizantal; Subsumption Issues; Layers, Allen; Language and AFSM, Layering and Networks; Module Indipendence and Sequencing; Communication through the World; Herbert; Subsumption Design; Genghis, the Nerd Herd; Tom and Jerry; Cambrian Intelligence; The Behavior Language User's Guide

Subsumption Architectures


26 Ottobre

Motor Schemas; Behaviors and Schema Theory; Architecture: perceptual and motor schemas; Output of MS: behavioral fusion; Primitive MS; Potential Fields; Issues on computing and combining PF; Local Minima; MS encoding: move-to-goal, avoid-static-obstacle, stay-on-path, move-ahead; noise, Schema design

Robot Architectures: Schema

27 Ottobre

Emergent behavior; Interaction: wall-following and flocking; Expectation and Observation; Uncertainty; Sequential vs. Parallel; Architecture and Emergence; Hybrid Control; Reusing Plans; Universal Plans; Reactive-Deliberative Trade-off; Architecture Classification

Robot Architectures: Emergence


2 Novembre

Locomotion Concepts; Found in Nature; Walking or Rolling; Issues; Legs and Joints; Gait; Walking Robots; Wheel Locomotion; Wheel types and arrangements; Examples

Robot Locomotion

3 Novembre

Small wheel-driven and legged Robot; Rhex and Wheg: three-spoke wheg leg; Wheg robot and gait; Jumping mini-wheg; Mobile robot kinematics: model and constraints; Common wheel applications; Differential drive robot; Degree of mobility; Maneurability; Degree of steerability; three-wheel configurations; Degree of Freedom; Holonomy

Wheel-legged Robot


22 Novembre

Bipedal walking: static and dynamic; Support phases and walking speed; Centers of Mass and Pressure; Inverted Pendulum; Walking Pattern; Inverted Pendulum force-driven: constant height travelling; Linear Inverted Pendulum: dynamic, behavior, transfer time, energy; Gait pattern: phases, energy; Linear Actuator: model, dynamics; Multi Joint Model; Adding Ankle Joint

Bipedal Locomotion: quasi-static gait

Behaviors: definitions and types; Reflexes; Taxes; Adaptive Behaviors; Behavior Interactions: Inhibitory and Cooperation; Arbitration; Robustness

Behavior-based Programming: Part One

23 Novembre

Bipedal Walking Pattern; LIP dynamics with a given Propulsive Force; walking on Ramps; LIP-3D Walking: properties and trajectory generation; Inverse Kinematics; Bezier Curves; Walking Pattern Generator: walking gait; Pattern Generation Issues; Walking Cycle: walk and turn patterns

Bipedal Locomotion: gait generation

Programming Behaviors: Wandering; Collision Avoidance; Escape; Homing; GPS Homing; Wall Following; Emergent Behaviors

Behavior-based Programming: Part Two

24 Novembre

Dynamical walking and stability; ZMP definition and computation; Legged Locomotion Issues; Trajectory-based methods: ZMP approach and stabilization; Minimalist Control: PID controller, Step Function, Hand-tuned, Open-loop Sinus-based; Heuristic Control: Virtual Model Control

Bipedal Locomotion: ZMP-based gait


29 Novembre

Some humanoid robots; What is legged locomotion; ZMP-based and limit-cycle-based; Passive dynamic walking; Hypothesis on legged locomotion; Mechanical design; Control methods; Role of sensor feedback; Stabilization of forward speed; Touchdown angle control; Hopping robots; Quadruped and Hexapod robots; Maneuverability; Legged robot issues; Neural system model control; Tuning of muscle tone; Stance/swing phases stiffness; CPG: neural oscillator, motion adaptation; long cyclic period walking; Reflex for rolling; Energy consumption

Robot Locomotion: biological inspiration

Anatomy of Running; Mechanics of Locomotion; Kinematics of Locomotion; Dynamics of Locomotion: inverted pendulum model, zero moment point

Biped Locomotion

30 Novembre

Locust behavior; Generation and Regulation of the motion; Neural model; CPG-fundamentals; Spinal Rhythm generation; The problem of legged locomotion control; Different types of gaits; statically vs. dynamically stable gaits; Minimalist control diagram; CPG-and-reflex-based control; Concept of limit cycle; Taga´s neuromechanical simulation; Non-linear-oscillators; Generation of traveling/standing waves for swimming/walking; Complex CPG model; The salamander model; CPG motor schema; CPG-and-reflex-based quadruped control; Motion generation and adaptation; Re-stepping response walking

Robot Locomotion: CPG-based locomotion

1 Decembre

Leg Function; Jumping for distance; Control of segmented leg; Generation of muscle activity; Running with elastic legs; from Running to Walking; Spring mass walking; Simple legged robots; Hip control; Leg segmentation and gait; Passive and Compliant walking

Robot Locomotion: emergence of gait


14 Dicembre

Terminology; Collective Systems: pros, cons, type; Application Domains; Individual and Collective Behaviors; Communication and Cooperation; Interference and Conflict; Competition; Predictive Models; Social Rules; Efficiency and Robustness; Performance

Multi Robot Systems: an introduction

Elastic Inverted Pendulum Model; Elastic IP Walker; Dynamic of the Elastic Pendulum; Balancing gravity force; Gait Cycle; Gait Pattern

Robot Locomotion: elastic gait

15 Dicembre

How group behaviors can be controlled: centralized, distributed, deliberative, hybrid, reactive, behavior-based; Approaches to group behavior; Foraging, Consuming, Mazing, Flocking, Pushing; Taxonomy: team size and structure, communication arrangement; Cao's, Yuti's, Parker's, Balk's, Dudek's Taxonomy; Examples; Stagnation; Box Pushing: task and implementation; Motion Coordination; Boids

Multi Robot Systems: control architectures


21 Dicembre

Objective of communication; The Nature of Communication; Communication Taxonomy and Issues; Range, contents, type of communication; Tasks for communication: experimentation and results; Task paramenters; Metrics; MRS performance evaluation; Diversity Measure; Simple and Hierarchic Social Entropy; Diversity Metric; Classification and Clustering

Multi Robot Systems: communication

22 Dicembre

Collective and group Behavior; Flocking and robot formation: herds, flocks and schools; ALLIANCE system: motivational behaviors; Cooperative transport: swarm-based and pray retrieval; Stagnation Recovery; Cooperative box-pushing; Agent definition and comparison; Autonomous agent; Colony of agents: social agent; Stigmergy; Individual and social system value; Social pressure and social behaviors; Macro-parameters and macro perceptions

Multi Robot Systems: collective robotics


11 Gennaio

Collective Robotics: taxonomy and benefits; Swarm Robotics and Intelligence; Natural Swarms: self-organization and emergence; Swarm Robotics: applications; Cooperation and Communications; Biological roots: bacteria, ants, wolves, primates and humans; Stigmergy and Swarms; Social Interactions; Simulations: U-bot experimentations; Cooperation Issues; Cooperative Transportation: pray retrival and group transport; Nestmate recruitment and coordination; Performance; Box Pushing; Issues in Swarm Robotics

Multi Robot Systems: swarm robotics

12 Gennaio

Ant Algorithms; Shortest Route; Visibility; Traveling Salesman Problem; Evaporation, Trail Intensity and Transition Probability; Open Questions; Cooperative Behavior vs. Role Allocation; Multi Robot System Issues: communication, cooperation, task planning; MRS Hybrid Architecture; Layered Levels of Control; Role Allocation in RoboCup; Utility functions; Master/Support Roles Coordination: role notification on the first notified/first advocated basis

Ant Systems and Role Allocation


18 Gennaio

Learning and Adaptation; Learning motivation in robotics; Why learning in robotics is hard; Learning types and methods; Reinforcement Learning; Thorndike's Law of Effect; RL Issues; Q Learning algorithm; Neural Networks; Classical conditioning; AHC and Neural Networks; Genetic Algorithms; Classifier Systems; Fuzzy Learning; Case-based learning; Memory-based learning; Explaination-based learning; Multistrategy learning

Learning in Robotics

19 Gennaio

Neural Networks: biological and artificial; How neurons work; AI and Cognitive Science; Neural Networks Issues: neural model and properties; Perceptrons; Thresholds; Learning in NN: perceptron learning rule; Linear separability; Perceptron Convergence Rule; Multilayer Perceptrons; Recurrent Networks; Derivative-based Optimization; Derivative-free Optimization; Backpropagation learning; Learning rate parameter; Neural Control: mimic, direct inverse, specialized inverse, back-propagation; Predictive Control

Robotics: Intelligent Control Methods