CSMARRT

Group 10

Cameron Kinard, Leaundre Zeno, Megan Wiedmaier, Heath Carley

Key Words: Neural Network, Creativity Machines, Robotic Markup Language, Perturbations

CSMARRT (Creative, Self-Learning, Multi Sensory, Adaptive, Reconfigurable, Robotics Toolbox) is a new neural technology being developed for the US military but also shows promise for civilian application. The software is designed to be used in robots giving them the ability to learn, develop, and adapt. These abilities will allow robots to carry out physical operations and overcome challenges that arise during the operation. CSMARRT utilizes Creativity Machine, a special form of a neural network, as a basis for problem solving. The software is in early development, but the preliminary tests have had promising results. As the software becomes more developed and tested in the military, it will be able to be transformed into a useable tool in the civilian world.

CSMARRT is being developed for the Air Force Research Laboratory by a company called Imagination Engines Inc. (IEI). IEI is a small company, founded by Stephen Thaler, specializing in synthetic intelligence. The project is driven by software created by Thaler. His software, known as Creativity Machine Paradigm, is what gives CSMARRT the ability to generate new ideas from existing information. This type of software is a new evolution of neural network used to produce an output function. The software is written in a new form of XML called Robotic Markup Language (RML) allowing designers to tailor the robot to a specific purpose. The software has two special features. The first feature special to neural networks is the use of perturbations to agitate existing information and arrive at new ideas. Perturbations are the source or trigger that stimulates the Creative Machine into new ideas. The second feature takes these new ideas and compare them to existing knowledge so that the unusable information can be filtered out and discarded. The software uses neural network paradigms as a learning tool. Creativity Machines learn from their mistakes and create and execute strategies to overcome these mistakes. The important issue as it applies to CSMARRT is that Creativity Machine neural networks are capable of generating solutions opposed to only recognizing patterns.

The ability to adapt will be key to the success of this project which is the strong point of the Creativity Machine. The project has already shown impressive results. The software has written music, discovered minerals, and designed soft drinks. In early testing, Thaler claims the software evaluates a situation and finds a quick solution to the task at hand. An example of this is when a test robot adopted a two-legged gait rising up on its hind legs so that it could move faster when necessary. In simulations the robots grouped together in order to overcome obstacles. They worked in teams to map their environment and strategically derived quick and effective ways to overcome these obstacles.

The military hopes to use this problem solving software in robots that can carry out particular missions ranging from projected scenario results to long-range tactical searches. The military wants to utilize the robots in missions similar to the operations in Afghanistan. The robots would be sent out to explore underground bunkers, map facilities, and identify what is in them. This is beneficial in many ways. One of the more important points is the ability to eliminate human involvement in very dangerous search missions, thus saving lives. The robots would also be able to go into areas that a human could not. The robots will be small in scale, measured in inches as opposed to feet. One possibility is the use of these robots to handle and dispose of mines. This technology could easily be applied to such a mission. This innovative new technology will allow the military to do things with robots that could not previously be done. CSMARRT gives the robots the flexibility to enter an area and deal with unknown obstacles without human control.

This software will first be used in the military but could easily be modified for commercial and industrial use. CSMARRT could potentially turn any robot into a valuable tool that is capable of performing a variety of tasks in the real world. This software could be adopted by many industries to assist in tedious physical tasks. The drilling industry is one example where CSMARRT could assist with the automation of everyday functions. It could be tailored to the industry to understand down hole scenarios. This could be a great asset to a drilling company if it possessed software that could forecast a particular well resulting in increased well cycles raising productivity and bringing maximized profits. It would be a huge competitive advantage. The problem solving capability of this software could eliminate the need for human involvement in dangerous activity. CSMARRT would allow industries to develop and implement new strategies. These strategies could increase the efficiency and productivity in many different environments.

CSMARRT is a new technology that has a wide variety of possible uses. The problem solving capabilities of this software is a new revolution in neural networks. The ability for software to be installed in a robot that can carry out specific functions while overcoming unknown obstacles could be invaluable in today’s fast-moving world. Even though it is still in the development stage it has shown promising results from preliminary testing. CSMARRT is being developed for the military but will soon be transformed into a useable tool for civilian use. It will be the new benchmark by which all other robotic software will be measured.

References

Hambling, David. (2006). Experimental AI powers robot army. Retrieved September 17, 2006, from http://www.wired.com/news/technology/software/0,71779-0.html?tw=wn_story_page_prev2

Imagination Engines Inc. (2005). IEI's patented Creativity Machine® paradigm. Retrieved September 17, 2006, from http://s129067403.onlinehome.us/cm.htm

Imagination Engines Inc. (2005). Robotic simulation environment. Retrieved September 17, 2006, from http://www.imagination-engines.com/robot_simulation.htm

Wikipedia. (2006). Neural network. Retrieved September 17, 2006, from http://en.wikipedia.org/wiki/Neural_network

CSMARRT
Group 10

1.	What is the name of the language used to write CSMARRT?

RML

RDL

DOL

CML

2.	What is a perturbation as it applies to the Creativity Machine Paradigm?

A.	The new idea that is reached through Creativity Machine software

B.	The classification of robots that the Creativity Machine software is used for

C.	The stimulant that triggers the process of obtaining a new idea

D.	Another name for the Creativity Machine Paradigm

1. 3.3. 2.3.

Who is the creative mind behind CSMARRT?

A.	Michael Rilee

B.	Dr. Ron McGaughey

C.	Owen Holland

D.	Stephen Thaler

4.	What is the name of the company that is developing CSMARRT?

A.	Robotics R Us

B.	Imagination Engines Inc.

ESSEX

IBOT

5.	What is the key to success for the Creativity Machine?

Marketing

B.	Gas Prices

C.	Adaptability

D.	Advertising

6.	The important difference between a Creativity Machine and other neural networks is?

A.	The ability to generate solutions

B.	Recognizing patterns

C.	Reconstructing learned memories

D.	None of the above

7.	Which of the following is NOT a benefit of CSMARRT?

A.	Increased productivity

B.	Eliminator of human involvement in dangerous activities

C.	Increased efficiency

D.	Ability to obtain knowledge uncontrollably