Rod Barman

Real-time control of soccer-playing robots using off-board vision: the dynamite testbed

The dynamite testbed has been constructed for experiments with autonomous mobile robots. It consists of a fleet of radio-controlled vehicles that receive commands from a remote computer. All the robots share an off-board vision system that tracks the position and orientation of each robot at 60 Hz. A novel aspect of this system is that computer vision is used not only for sensing the environment, but also for sensing the state of the robot itself. The former information is used for planning and the latter for control. The testbed has been used successfully for experiments with robot soccer games to compare the effectiveness of alternate control architectures. Testbed use is facilitated through a graphical simulator that allows control programs to be tested off-line at a workstation. The testbed has been developed to the point where we now have a portable version that has been demonstrated at a number of conferences.

Dynamo: Real-time Experiments With Multiple Mobile Robots

This paper describes the Dynamo project in which an extensive facility for experimental mobile robotics has been developed. It currently consists of nine radio-controlled mobile robots, two CCD color video cameras, a video transmitter and tuner, radio controllers, Datacube image processing hardware, and a network of transputers. Software for tracking, control, and simulation has also been developed.

Spinoza: a stereoscopic visually guided mobile robot

Our mobile robot, Spinoza, embodies a sophisticated real-time vision system for the control of a mobile robot in a dynamic environment. The complexity of our robot architecture arises from the wide variety of tasks that need to be performed and the resulting challenge of coordinating multiple distributed, concurrent processes on a diverse range of processor architectures, including transputers, digital signal processors and a workstation host. The system handles the sensing, reasoning and action components of a robot, distributed over these architectures, and responds to unpredictable events in an unknown dynamic environment. Spinoza relies heavily on its capability to perform real-time vision processing in order to perform task such as mapping, navigation, exploration, tracking and simple manipulation.

SILT: a distributed bit-parallel architecture for early vision

A new form of parallelism,distributed bit-parallelism, is introduced. A DBP organization distributes each bit of a data item to a different processor. DBP allows computation that is sublinear with word size for such operations as integer addition, arithmetic shifts, and data moves. The implications of DBP for system architecture are analyzed. An implementation of a DPB architecture based on a mesh with a fost-bypass network is presented, and the performance of DBP algorithms on this architecture is analyzed. The application of the architecture to early vision algorithms is discussed.

Silt: the bit-parallel approach

A particular form of parallelism, called bit-parallelism, is introduced. A bit-parallel organization distributes each bit of a data item to a different processor. Bit-parallelism allows computation that is sublinear with word size for such operations as integer addition, arithmetic shifts, and data moves. The implications of bit-parallelism for system architecture are analyzed. An implementation of a bit-parallel architecture based on a mesh with a bypass network is presented. Using a conservative estimate for cycle time, a Silt processor performs 64-b integer additions more than 10 times faster than the Connection Machine-2. Using current CMOS technology, a 16 M processor Silt system would be capable of nearly 500 billion 32-b adds per second. The application of the architecture to low-level vision algorithms is discussed.<<ETX>>

State assignment for multilevel logic using dynamic literal estimation

The problem of assigning a representation for the states in a finite-state machine (FSM) so as to minimize the area required for implementation using multilevel logic is considered. The problem is reduced to one of minimizing the literal count in the combinational logic part of the FSM. Heuristics are used to guide the selection of candidate state assignments, and a fast literal estimator is used to choose the best assignment from among the candidates. This approach is compared to the mustang program and is found to produce, on average, smaller literal counts than mustang.<<ETX>>