Our work has been guided and influenced by the many fine readings assigned in class. To wit:

A Brief List of Fine Robotics Papers Which Influenced Us

Achieving Artificial Intelligence through Building Robots, by Rodney Brooks

Advises researchers to build "robotic insects" which don't make explicit plans.

The Polly System, by Ian Horswill

Describes a tour-giving robot designed by along the lines advocated in the paper above.

Experiments in Automatic Flock Control, by Richard Vaughan et al

A Robotic Sheepdog that gathers a flock of ducks

DERVISH: An Office-Navigating Robot, by Illah Nourbakhsh et al

A robot that uses states and uncertainties in order to determine its location

Monte Carlo Localization for Mobile Robots, by Frank Dellaert et al

Describes Monte Carlo Localization and details its navigational capability

Probabilistic Robot Navigation in Partially Observable Environments, by R. Simmons and S. Koenig

Another very nice paper on navigating using only local information.

Robot Evidence Grids, by M. C. Martin and H. Moravec

Moravec strikes again. More statistical reasoning for robot navigation.

Subconsciously, Athletes May Play Like Statisticians, by D. Leonhardt

Suggests that human athletes are actually using a Bayesian approach to decision making. A very interesting idea.

Coastal Navigation: Robot Motion with Uncertainty, by N. Roy, W. Burgard, D. Fox, S. Thrun

Coastal navigation is a technique that keeps the robot in places which are inherently most localizable. The idea is to keep the robot in a high entropy environment, so that is sensors will be maximally useful, while still proceeding towards a goal.

Millibots: The Development of a Framework and Algorithms for a Distributed Heterogenous Robot Team, by L. Navarro-Serment, et al.

Describes techniques for mapping, sensing, and navigation of many autonomous robots with communication between them. Teams of robots can be much more effective than individual robots operating in parallel.

Walk on the Wild Side: Designers of the PolyBot Robot System Solve the Challenges of Locomotion by Mimicking Locomotion in the Animal World, by M. Yim, et al.

PolyBot is a robotic system made of many modules, which can be arranged in a variety of ways. The flexibility afforded by this reconfigurability allows the robot to take many forms and use different methods of propulsion. Among those forms are that of the snake, the ring, and a four-legged spider, which allow it to go up stairs, vertical mesh surfaces, and travel quickly on flat surfaces.

RRT-Connect: An Efficient Approach to Single-Query Path Planning, by J. Kuffner Jr. and S. LaValle

Presents a randomized algorithm for solving path planning problems in configuration spaces.