COGSCI 1 Lecture 18 - Robotics

GOFAI

Good Old-Fashions Artificial Intelligence Robotics

Shakey

• Early robot developed by Stanford Research Institute (c. 1970)

• First robot that was able to move around, perceive, follow instructions, and implement complex instructions

• Software on a separate computer system and communicated withe Shakey via radio antenna

Limitations of GOFAI

the robot is not embedded in a real-life environment and can never really come to terms with real-life problems and challenges;

• Can only operate in a highly constrained environment

• Cannot learn to solve problems; solutions are built in

Dynamical Systems

behaviors emerge out of complex interactions between an organism and its environment

Traditional explanation for U-shaped developmental trajectory of stepping

• Infant’s initial stepping movements are purely reflexive

• They disappear during the non-stepping window bc the cortex has matured enough to inhibit reflex responses

Studies by Esther Thelen and Linda Smith

research indicated that stepping movements could be artificially induced or inhibited in infants by manipulating features of the environment

• Placing baby on treadmill or suspending them in warm water

→ Stepping movements vary independently of how the cortex has developed

Dynamical Models

• Are used to understand how agents are embedded in their environments

• Use calculus-based methods to track the evolving relationship between a small number of variables over time

Situated Cognition

• Propose a dynamical systems-like approach to robotics

• Believe that we should start small and focus on basic ecologically valid problems

Barbara Webb Robot Crickets

Can identify the source of a sound and move automatically toward the source without any of the systems normally assumed by GOFAI

Biorobotics

Using knowledge of living insects, as well as AI, to create agents capable of moving about and solving problems in their environment

Morphological Computation

exploiting features of body shape to simplify what might otherwise be highly complex information-processing tasks

Morphological Computation in Robotics

building as much of the computation as possible directly into the physical structure of the robot

Yokoi Hand

Constructed from elastic and deformable materials that allow the hand to adapt itself to the shape of the objects being grasped

Subsumption Architectures

Operate on a set of relatively simple stimulus-response mechanism

• Bottom up intelligence

• Based on the idea that intelligence does not require formal symbolic representation

• Robots make reflexive responses to environmental stimuli

Obstacle Avoidance Layer

Directly connects perception to action

Rodney Brook’s Robot Allen

• Basic layer = obstacle-avoidance layer

• More layers were added overtime, mimicking evolution

• Semi-autonomous subsystems operate independently of each other

• No central “controller” comparable to PLANEX in SHAKEY

• Direct perception-action links allows robot to deliver immediate motor responses

Situated Cognition

embedded in the world, and which does not deal with abstract descriptions, but through its sensors with the here and now

Embodied Cognition

has a physical body and experiences the
world directly through the influence of the world on that body

Xenobot

• Small Biological Machine (<1mm) created by scientists at the University of Vermont and Tufts University

• Built using skin and heart cells harvested from frog embryos

• Designed and programmed by a supercomputer using an evolutionary algorithm

• Able to move in a coherent fashion to explore their watery environment
  and can survive for days or weeks, powered by embryonic energy stores

Imitation Learning

Roboticist puppeteers pairs of metallic robotic arms;

• helps master fine motor movements of hands: one of the biggest challenges in robotics

Reinforcement learning

subdivides each demonstration into a series of sub-tasks and mines successful simulations

→ able to learn more from less data

Affective Computing

Computing that relates to, arises from, or deliberately influences emotions

Uncanny Valley Effect

humanoid objects that imperfectly resemble actual human beings provoke uncanny feelings of uneasiness and revulsion in observers

(eg. Polar Express)

Moley Roboics

robot kitchen that cooks and cleans up after itself

Automated Restaurants

• Spyce Restaurant in Boston: created by a team of MIT graduates

• CaliExpress by Flippy in Pasadena is first AI-powered eatery

• Qianxi Robot Catering Group has introduced fully automated restaurant where all processes from cooking to serving are handled by robotic machines

Advantages of Autonomous Restaurants

• Every second added to someone waiting for their meal is lost opportunity for revenue for restaurant, so speed of automation at these restaurants is a big plus

• Prices at these restaurants tend to be very competitive

• Can produce 300 bowls of food an hour

• Can sear proteins, steam grains, and measure toppings to create delicious food in less than 3 min

• Also, the food will presumably be cooked precisely the same every time so the automation “should result in improvement in taste”

• Customers tend to love robot waiters