Interesting AI Demos and Projects
building autonomous agents to simulate group motion and obstacle avoidance
such as activities of bird flocks and fish schools
The project is to develop a generic architecture for a group of agents
to pursue their given goals, adapt their behavior to new environments, and
communicate and perform coordinated group actions.
Agents work at IBM
Video Environment (MIT)
The Interactive Video Environment (IVE) is an experimental "testbed"
to explore how Computer Vision and Computer Agents technologies can be used
to solve problems of human interface and human interaction over networks.
The goal of the IVE project is to develop smart cooperative work/play
systems that function robustly despite wide variation in network and
Softbots (U. Washington)
Building autonomous agents that interact with real-world software
environments such as operating systems or databases is a pragmatically
convenient yet intellectually challenging AI problem. We are utilizing
planning and machine-learning technology to develop an Internet softbot
(software robot), a customizable and (moderately) intelligent assistant for
Internet access. The softbot accepts goals in a high-level language,
generates and executes plans to achieve these goals, and learns from its
- Guardian Angel
The project uses guardian angels(software agents) to create health
information systems centered on the patients rather than solely for the
convenience of the doctors.
- Kasbah (MIT)
Kasbah is an an on-line multiagent classified ad system. Users can create
agents that, being given certain directions, strategies, and constraints,
are launched into a common "marketplace" where they seek out and
negotiate with, potential buyers or sellers.
Resources on Agents
Intelligent Web Apps
Game Design and Playing
Hal is a highly interactive environment that uses embedded computation to
observe and participate in the normal, everyday events occurring in the
world around it. It has cameras for eyes, microphones for ears, and uses a
variety of computer vision, speech and gesture recognition systems to allow
people to interact naturally with it.
PAT: A Virtual Personal Aerobics Trainer(MIT)
A prototype system for implementing a virtual Personal Aerobics Trainer
(PAT) is presented. This system allows the user to create and personalize an
aerobics session to meet the user's needs and desires. Various media
technology and computer vision algorithms are used to enhance the
interaction of the character by enabling it to watch and talk to the user.
- Autonomous Vehicle Navigation using Neural Nets (CMU)
ALVINN uses neural networks to learn visual servoing. It watches a person
drive for five minutes, and can then take over driving. ALVINN has been
trained to drive on dirt paths, single-lane country roads, city streets, and
multi-lane highways. Click here
for images of the vehicles and videos of ALVINN in action. The sucessor to
ALVINN, called RALPH,
was the core of a system that drove a vehicle autonomously all but 52 of the
2,849 miles from Pittsburgh to San Diego, averaging 63 miles per hour, day
and night, rain or shine.
Lisp Hypermedia Server(MIT)
This server is created with Lisp. It has inductive learning ability and uses
natural language processing techniques to answer questions.
ICE Neural Nets Hot List has intro material on artificial neural
networks, as well as some Java demonstrations
an integrated workbench for the construction of adaptive knowledge-based
systems and data mining using first-order ML techniques (ILP).
- NeuroOn-Line Producta
complete graphical, object-oriented software tool kit for building neural
network applications and applying them to dynamic environments.
The VISION Projec(Vibration Interpretation using Simulation and the
Intelligence of Networks) utilizes the incorporation of neural networks,
knowledge-based systems, and other techniques.
Identification using a Decision Tree
Natural Language Processing
Intelligence and Robotics Project(AIR) in Italy(Politecnico di Milano)
The Cog Shop builds, maintains, and experiments with Cog, a humanoid robot.
II Walking Robot (CMU)
The CMU Field Robotics Center (FRC) developed Dante II, a tethered walking
robot, which explored the Mt. Spurr (Aleutian Range, Alaska) volcano in July
1994. The use of robotic explorers, such as Dante II, opens a new era in
field techniques by enabling scientists to remotely conduct research and
of two robot motion planning algorithms(University of Minisota)
A Robot for Social Interactions with Humans(MIT)
The Robotic Tour Guide (CMU and University of Bonn)
Minerva is an intelligent mobile robot tour-guide that moves daily through
crowds at the Smithsonian's National Museum of American History. You can
know more about it from its
homepage at CMU.
Robot Navigation using Stereo Vision (JPL)
The NASA JSC Mobile Robot Lab recently demonstrated the integration of a
stereo vision system with a mobile robot for the purpose of following people
or other robots. During following, the stereo system is constantly feeding
updated coordinates of the agent being tracked to the mobile robot system.
The mobile robot takes these coordinates as goal positions and attempts to
attain the goal position while avoiding obstacles using its sonar sensors.
The MERCURY PROJECT allows users to tele-operate a robot arm moving over a
terrain filled with buried artifacts. A CCD camera and pneumatic nozzle
mounted on the robot allow users to select viewpoints and to direct short
bursts of compressed air into the terrain. Thus users can
"excavate" regions within the sand by positioning the arm,
delivering a burst of air, and viewing the newly cleared region.
- Tao-1 Intelligent
Wheelchair (AAI Inc.)
The infared sensors are used in this intelligent autonomous wheelchair to
collect the data of enviornment. After the data processing, the wheelchair
can avoid the obstacles automatically and navigate based on the landmark.
and Grasping Moving Objects (Columbia)
Coordination between an organism's sensing modalities and motor control
system is a hallmark of intelligent behavior, and we are pursuing the goal
of building an integrated sensing and actuation system that can operate in
dynamic as opposed to static environments. The system we are building is a
multi-sensor system that integrates work in real-time vision, robotic arm
control and stable grasping of objects. Our first attempts at this have
resulted in a system that can track and stably grasp a moving model train in
Robotics using estimation(Northwestern AI Research Lab)
theorem prover proved a long-standing mathematical conjecture about
algebra, called the Robbins Problem
Miscellaneous AI-aided applications