ControlShell: A Real-Time Software Framework.
Abstract
We have created a new paradigm for building and maintaining complex real time software systems for the control of moving mechanical systems. This objective has been met through the simultaneous development of both a powerful software environment and cogent motion planning and control capabilities. Our research has concentrated on three key areas: (1) Building an innovative, powerful real time software framework; (2) Implementing new distributed control architectures for intelligent mechanical systems; and (3) Developing distribution architectures and new algorithms for the computationally 'hard' motion planning and direction problem. This research built on ControlShell, originally developed at the Stanford Aerospace Robotics Laboratory and now marketed by Real-Time Innovations, Inc. The research added fundamental new capabilities, including network-extensible data flow control to allow scalable support for distributed systems and a graphical environment to enable rapid prototyping and increased productivity. The concurrent development of this tool and experimental robotic applications was aimed at ensuring high quality architectural design and producing reusable components. The research was also aimed at developing more specific modules, mainly motion planning modules which can operate under time constraints. Since motion planning is a computationally hard problem, one cannot expect to achieve strict real time motion planning.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jul 01, 1995
- Accession Number
- ADA327565
Entities
People
- Jean-claude Latombe
- Robert Cannon
- Stan Schneider
Organizations
- Stanford University