Kinematic Self-Replicating Machines

© 2004 Robert A. Freitas Jr. and Ralph C. Merkle. All Rights Reserved.

Robert A. Freitas Jr., Ralph C. Merkle, Kinematic Self-Replicating Machines, Landes Bioscience, Georgetown, TX, 2004.


5.8 Software Simulators for Robots and Automated Manufacturing

Valentí Pineda [1314], an industrial robotics and automation engineer, notes that in common practice the use of accurate software simulators can save a lot of time, money and effort in a project, and also “are sometimes a powerful argument to convince a customer to make an investment when they see a virtual factory performing virtual work with their name written somewhere in the working cell.”

Some of the benefits of simulators in industrial robotics may be applicable to robotic/automated replicators, including: (1) calculation of the optimum layout that favors industrial processes needed for the fabrication of the replicated unit; (2) calculation and verification of job concept before undertaking the real work, allowing a determination of which operations are possible and which ones are not; (3) optimizing the design of devices, pieces and sub-assemblies for most effective grasping, recognition and reaching the position effectively by robots; (4) deciding how many robots and devices are required to successfully complete the operations; (5) optimization of required resources prior to the hardware implementation; and (6) operational testing of all the above to verify and overcome positioning problems [1314].

Besides robotics simulators, many other types of software may help in the design and testing of replicators. For example, 3-D drawing software (e.g., SolidWorks, AutoCAD (AutoDesk), 3D Studio Max, etc.) allows the engineer to see what the best assembly process should be and to simulate pieces in motion. The 3-D files can be exported back to robotic simulation software (e.g., CimStation Robotics [2819], Deneb/IGRIP (Interactive Graphics Robot Instruction Program [2820]), MotoSim [2821], WorkSpace [2822], and mobile robotics simulators [2823]), using which various aspects of robotic manufacturing operations can be tested – including kinematics, dynamics, motion planning, verification of production concepts, workcell designs, and manufacturing processes – prior to physical implementation. Design for Assembly software tools facilitate development of multilevel assemblies, sequences, part paths and process documentation [2824]. Other types of software permit testing of wear, thermal stress, mechanical stress, and so forth. The use of visual programming languages (e.g., Visual Basic or Visual C++) also permits the simulation of chemical or physical processes implementing equations in the program and showing a visual output of the processes, such as the simulation of an expert system controlling a chemical reactor [1314]. General purpose manufacturing process simulation packages such as Flexsim [2825], ShowFlow [2826] and SimCAD [2827] are well known.

At the high end of complex machine design, we have, for example, the comprehensive 3-D modeling tools of Dassault Systemes, which Boeing used in the early 1990s to create the world’s first entirely digitally designed aircraft (thus avoiding the expense of creating physical prototypes). Subsequent generations of this software are now being used to implement “product lifecycle management,” suggesting a coordinated process by which thousands of engineers located around the world might, in the future, cooperate in the design of a single complex nanorobotic system containing millions of parts and billions of atoms. Explains one writer [2996], of the Boeing effort: “If the supplier of a section of the fuselage, say, decides to move the location of a door by 2 inches, that will affect the design of everything around it, from the placement of windows to the location of hidden wiring bundles. In the past such a change in the door would require weeks to redesign adjoining parts. Dassault’s software instead will allow designers anywhere to instantly know how a modification in their piece of the [aircraft] will affect every other part of the plane. The software will also allow Boeing to simulate the way the plane will be assembled in the factory before a single piece of machinery is purchased, [avoiding] costly delays if engineers need to rejigger factory bottlenecks. The software can also simulate wear and tear over many years to predict, for instance, when a hydraulic pump is likely to fail.”


Last updated on 1 August 2005