Silvano P. Colombano

• Doctor of Philosophy
• Retired at National Aeronautics and Space Administration

There is no fundamental reason why A-life couldn't simply be a branch of computer science that deals with algorithms that are inspired by, or emulate biological phenomena. However, if these are the limits we place on this field, we miss the opportunity to help advance Theoretical Biology and to contribute to a deeper understanding of the nature of...

Autonomy is required for manned spacecraft missions distant enough that light-time communication delays make ground-based mission control infeasible. Presently, ground controllers develop a complete schedule of power modes for all spacecraft components based on a large number of factors. The proposed architecture is an early attempt to formalize an...

NASA's exploration program envisions the utilization of a Deep Space Habitat (DSH) for human exploration of the space environment in the vicinity of Mars and/or asteroids. Communication latencies with ground control of as long as 20+ minutes make it imperative that DSH operations be highly autonomous, as any telemetry-based detection of a systems p...

The Credibility Assessment (CA) process is designed to give project decision makers a detailed picture of the quality of the results obtained from simulations from several possibly crucial points of view. This allows them to better understand the risks that might be associated with the use of those results in the context of design and/ or process d...

This paper reviews the range of technologies that will contribute to the construction of space infrastructure that will both enable and, in some cases, provide the motivation for space exploration. Five parts are addressed: Managing complexity, robotics based construction, materials acquisition, manufacturing, and self-sustaining systems. © 2010 by...

Identifying DNA splice sites is a main task of gene hunting. We introduce the hyper-network architecture as a novel method for finding DNA splice sites. The hypernetwork architecture is a biologically inspired information processing system composed of networks of molecules forming cells, and a number of cells forming a tissue or organism. Its learn...

To increase the quality of scientific data collected from autonomous mobile agents such as rovers and walking robotic devices, biological methods can be mimicked for better navigation and balance control of both the agent itself and the manipulation of scientific instruments. Drawing on the design of the neuro-vestibular control system, the EarBot...

Without Abstract

The Scorpion robot is an innovative, biologically inspired 8-legged walking robot. It currently runs a novel approach to control, which utilizes a central pattern generator (CPG) and local reflex action for each leg. From this starting point we are proposing to both extend the system's individual capabilities and its capacity to function as a "scou...

A technology of self-sustaining robotic ecologies (called "robosphere" by the co-author) is likely to emerge in order to prepare for and serve long- duration space missions of humans on Mars or in other extraterrestrial environments. The emergence of this would necessarily expand the concept of the human factors. The human-human, human- technology,...

The goal of an evolutionary algorithm (EA) is to find the global optimum in a state space of potential solutions. But these systems can become trapped in local optima due to the EA having only generational information. Using the scouting algorithm (SA) it is suggested that a cross-generation memory mechanism can be added to modulate fitness relativ...

Contents include the folowing: rom "one shot" explorations to infrastructure building. Challenges to infrastructure building. Modularity and self-sustaining robotic ecologies. A pathway to human presence. Robotic " archntecture". The "robosphere" concept.

The hypernetwork architecture is a biologically inspired learning model based on abstract molecules and molecular interactions that exhibits functional and organizational correlation with biological systems. Hypernetwork organisms were trained, by molecular evolution, to solve N-input parity tasks. We found that learning improves when molecules exh...

NASA needs autonomous robotic exploration of difficult (rough and/or steep) scientifically interesting Martian terrains. Concepts involving distributed autonomy for cooperative robotic exploration are key to enabling new scientific objectives in robotic missions. We propose to utilize a legged robot as an adjunct scout to a rover for access to diff...

The present sequential mission oriented approach to robotic planetary exploration, could be changed to an infrastructure building approach where a robotic presence is permanent, self sustaining and growing with each mission. We call this self-sustaining robotic ecology approach robosphere and discuss the technological issues that need to be address...

Experimental exploration of enzymatic response in a multidimensional context faces the challenge of an explosive number of possible milieu conditions. We address this problem with an evolutionary search strategy that scouts the physiochemical milieu space for unanticipated enzyme behavior and rewards the discovery of experimental conditions that yi...

We describe a parallel genetic algorithm (GA) that automatically generates circuit designs using evolutionary search. A circuit-construction programming language is introduced and we show how evolution can generate practical analog circuit designs. Our system allows circuit size (number of devices), circuit topology, and device values to be evolved...

ample: "bricks" (Penrose and Penrose, 1957), model railroad cars (Jacobson, 1958), and floating electromagnets (Morowitz, 1959). In addition to being clever, these designs could be tested and demonstrated through physical models -- making them compelling and amenable to direct investigation. Designing a self-assembling structure is akin to engineer...

This paper addresses the origin of robust and evolvable metabolic functions, and the conditions under which it took place. We propose that spatial considerations, traditionally ignored, are essential to answering these important questions in prebiotic evolution. Our probabilistic cellular automaton model, based on work on autocatalytic metabolisms...

Yagi-Uda antennas are known to be di#cul to design and optimize due to their sensitivity at high gain, and theincl#wY4 of numerous parasitic elVYy ts. We present a genetic alzyzw4Iw#VWy4 automated antenna optimization system that uses a fixed Yagi-Uda topolwY and a byte-encoded antenna representation. The fitnesscals4YWBWw al l ws the implxBB relBB...

High-level analog circuit design is a complex problem domain in which evolutionary search has recently produced encouraging results. However, little is known about how to best structure evolution for these tasks. The choices of circuit representation, fitness evaluation technique, and genetic operators clearly have a profound e#ect on the search pr...

We present a method of automatically generating circuit designs using evolutionary search and a set of circuit-constructing primitives arranged in a linear sequence. This representation has the desirable property that virtually all sets of circuit-constructing primitives result in valid circuit graphs. While this representation excludes certain cir...

Robotic exploration of Mars has been a "one shot" approach where each surface mission is planned typically with a rover that will perform a series of experiments for a few weeks or months, until the robot becomes unable to operate in the harsh Mars conditions and simply "dies". It would clearly be desirable to have robots on Mars that can last for...

The fundamental problem in the evolution of matter is to understand how structure-function relationships are formed and increase in complexity from the molecular level all the way to a genetic system. We have created a system where structure-function relationships arise naturally and without the need of ad hoc function assignments to given structur...

Yagi-Uda antennas are known to be difficult to design and optimize due to their sensitivity at high gain, and the inclusion of nu- merous parasitic elements. We present a genetic algorithm-based auto- mated antenna optimization system that uses a fixed Yagi-Uda topology and a byte-encoded antenna representation. The fitness calculation al- lows the...

We present a method of evolving analog electronic circuits using a linear representation and a simple unfolding technique. While this representation excludes a large number of circuit topologies, it is capable of constructing many of the useful topologies seen in hand-designed circuits. Our system allows circuit size, circuit topology, and device v...

The hypernetwork is a molecular interaction-based model that has learning capabilities. The adaptive algorithm randomly changes the molecular structures and selects the best individual. Experiments with the hypernetwork show the importance for evolution of the mutation buffering capabilities of the system. Mutation buffering allows the system to im...

Intelligence cannot be understood - and cannot be 'artificially recreated' - without also understanding how it fits as a phenomenon in the evolution of matter. I say 'matter' instead of 'life' because my thesis is that the distinction between matter, with its associated interactions, and life, is simply in the complexity of these interactions and i...

A philosophical view is given of artificial intelligence. The article examines: the behaviorist model; symbolic AI; intelligence and evolution; and cognition. It is concluded that the common framework for AI will have to come from brain architecture. Abstracting relevant principles remains extremely difficult until we really understand the interpla...

There is no fundamental reason why A-life couldn't simply be a branch of computer science that deals with algorithms that are inspired by, or emulate biological phenomena. However, if these are the limits we place on this field, we miss the opportunity to help advance Theoretical Biology and to contribute to a deeper understanding of the nature of...

The relatively new field of Evolvable Hardware studies how simulated evolution can reconfigure, adapt, and design hardware structures in an automated manner. Space applications, especially those requiring autonomy, are potential beneficiaries of evolvable hardware. For example, robotic drilling from a mobile platform requires high-bandwidth control...

The goal of this research is to investigate the application of evolutionary search to the process of automated engineering design. Evolutionary search techniques involve the simulation of Darwinian mechanisms by computer algorithms. In recent years, such techniques have attracted much attention because they are able to tackle a wide variety of diff...

We present a method of automatically generating circuit designs using evolutionary search and a set of circuit constructing primitives arranged in a linear sequence. This representation has the desirable property that virtually all sets of circuit-constructing primitives result in valid circuit graphs. While this representation excludes certain cir...

This paper explores a coevolutionary approach applicable to di#cult problems with limited failure#success performance feedback. Like familiar #predator-prey" frameworks this algorithm evolves two populations of individuals # the solutions #predators# and the problems #prey#. The approach extends previous work by rewarding only the problems that mat...

This paper addresses the origin of robust and evolvable metabolic functions, and the conditions under which it took place. We propose that spatial considerations, traditionally ignored, are essential to these important questions in prebiotic evolution. Our probabilistic cellular automaton model (inspired by work on autocatalytic metabolisms by Eige...

High-level analog circuit design is a complex problem domain in which evolutionary search has recently produced encouraging results. However, little is known about how to best structure evolution far these tasks. The choices of circuit representation, fitness evaluation technique, and genetic operators clearly have a profound effect on the search p...

A novel data processing algorithm, which addresses the inverse problem of deducing internal network structure from behavior, as represented by time-series data on signals at one or more output nodes, has been developed. Data analysis techniques, based on information theory and Bayesian change-point determination methods originally developed for ast...

We construct simple artificial chemistries in order to gain an understanding of how a chemical reaction network might emerge from a state of relative disorder in non-living “protocells”. Such chemistries have relevance to origin-of-life studies as well as to artificial life research. We present a model comprised of interacting polymers, and specify...

We present a method of evolving analog electronic circuits using a linear representation and a simple unfolding technique. While this representation excludes a large number of circuit topologies, it is capable of constructing many of the useful topologies seen in hand-designed circuits. Our system allows circuit size, circuit topology, and device v...

We model an artificial chemistry comprised of interacting polymers by specifying two initial conditions: a distribution of polymers and a fixed set of reversible catalytic reactions. A genetic algorithm is used to find a set of reactions that exhibit a desired dynamical behavior. Such a technique is useful because it allows an investigator to deter...

This paper details flight results from the PI-in-a-Box project. The project is an attempt to understand the issues involved in building automated real-time advisory systems to improve the conduct of science in remote laboratory settings. This is an important problem as examples of these situations can be found throughout aerospace. A knowledge-base...

Goal Directed Model Inversion (GDMI) is an algorithm designed to generalize supervised learning to the case where target outputs are not available to the learning system. The output of the learning system becomes the input to some external device or transformation, and only the output of this device or transformation can be compared to a desired ta...

Goal Directed Model Inversion (GDMI) is an algorithm designed to generalize supervised learning to the case where target outputs are not available to the learning system. The output of the learning system becomes the input to some external device or transformation, and only the output of this device or transformation can be compared to a desired ta...

NASA's space science program is based on strictly pre-planned activities. This approach does not always result in the best science. We describe an existing computer system that enables space science to be conducted in a more reactive manner through advanced automation techniques that have recently been used in SLS-2 October 1993 space shuttle fligh...

Goal Directed Model Inversion (GDMI) is an algorithm designed to generalize supervised learning to the case where target outputs are not available to the learning system. The output of the learning system becomes the input to some external device or transformation, and only the output of this device or transformation can be compared to a desired ta...

Experience with the design and implementation of a portable computing system for STS crew-conducted science is discussed. Principal-Investigator-in-a-Box (PI) will help the SLS-2 astronauts perform vestibular (human orientation system) experiments in flight. PI is an interactive system that provides data acquisition and analysis, experiment step re...

This paper describes portions of a novel system called MARIKA (Model Analysis and Revision of Implicit Key Assumptions) to automatically revise a model of the normal human orientation system. The revision is based on analysis of discrepancies between experimental results and computer simulations. The discrepancies are calculated from qualitative an...

The Principal Investigator (PI) in-a-Box knowledge-based system (KBS) helps astronauts perform science experiments in space. These experiments are typically costly to devise and build, and often difficult to perform. Further, the space laboratory environment is unique, ever-changing, hectic, and therefore stressful. The environment requires quick,...

A new neural network technique for model inversion called goal directed model inversion (GDM) is presented. It allows the system to produce an inverse model in a goal directed manner. The major advantage of an inverse model created in this matter is that it can adapt to unexpected changes in the system with which it must interact. As an example of...

Principal Investigator (PI) in a box is a computer system designed to help optimize the scientific results of experiments that are performed in space. The system will assist the astronaut experimenters in the collection and analysis of experimental data, recognition and pursuit of 'interesting' results, optimal use of the time allocated to the expe...

This paper describes an ongoing expert system development effort started in 1988 which is evaluating both CLIPS- and LISP- based approaches. The expert system is being developed to a project schedule and is planned for flight on Space Shuttle Mission SLS-2 in 1992. The expert system will help astronauts do the best possible science for a vestibular...

Work being done on an expert system designed to improve the quality of space experimentation in vestibular physiology is described. Emphasis is given to two of the several modules of which the system architecture is composed: the Protocol Manager and the Interesting Data Filter. Some engineering issues related to human factors and to data acquisiti...

Simulation and Artificial Intelligence share a fertile common ground both from a practical and from a conceptual point of view. Strengths and weaknesses of both Knowledge Based System and Modeling and Simulation are examined and three types of systems that combine the strengths of both technologies are discussed. These types of systems are a practi...

The methodology required for the development of the 'intelligent system architecture' of distributed computer systems which integrate standard data processing capabilities with symbolic processing to provide powerful and highly autonomous adaptive processing capabilities must encompass three elements: (1) a design knowledge capture system, (2) comp...

The version of DPNS Version 2 which is described in this document consists of two programs, the DPNS model and the report generator (sometimes referred to as the 'back end'). The DPNS Version 2 user interface, which is currently not available, is the third program in the package. It provides the user with an interactive tool for defining and valida...

The Gynecologic Oncology Group (GOG) is a multiinstitutional, multidisciplinary cooperative group. Its clinical trials encompass the planning, initiation, execution, analysis, and presentation of results of management and treatment of patients with gynecologic cancer. The GOG Statistical Office, located at Roswell Park Memorial Institute, is respon...

The problem of constructing life support systems which require little or no input of matter (food and gases) for long, or even indefinite, periods of time is addressed. Natural control in ecosystems, a control theory for ecosystems, and an approach to the design of an ALSS are addressed.

Autonomous Life Support Systems (ALSS) are envisioned for long range permanence in space. ALSS would require little or no input of matter for extended periods of time. The design of such a system involves an understanding of both ecological principles and control theory of non-linear, ill-defined systems. A distinction is drawn between ecosystem su...

The term „cascade spread,” with regard to metastases, is being used ambiguously. Some workers take it to mean that cancers metastasize along preferred pathways, others infer that some sites are seeded first and that further metastases proceed from these sites rather than from the primary tumor. It is our opinion that the statistical arguments used...

The most promising and efficient approach to discovery of improved modes of therapy in cancer involves multi-disciplinary studies in which basic scientists collaborate with their clinical brethren in the evaluation of promising leads suggested by laboratory studies. In this fashion, new ideas can be tested as quickly as possible and multivariate me...

A method for the calculation of helical parameters from internal coordinates of an arbitrary repeat unit, has been implemented in a FORTRAN IV program. The coordinates of all atoms in a cylindrical coordinate system are also obtained. An application of the program for a systematic approach to the problem of protein-DNA recognition is introduced.

Prior to the development of the CSFS, the logistics of hemoccult stool screening required many hours of clerical and secretarial time. In order to conduct multiple county, multiple location screening programs, the CSFS has been developed as a specialized application of the Roswell Park Management System (RPMIS). CSFS uses the RPMIS for formatted vi...

Computer model-building procedures using stereochemical principles together with theoretical energy calculations appear to be, at this stage, the most promising route toward the elucidation of DNA-protein binding schemes and recognition principles. A review of models and bonding principles is conducted and approaches to modeling are considered, tak...

Self-assembling systems are of great interest in nanotechnology. Although the design of such systems can differ widely at nano and macro scales, certain design principles are invariant. Because of these similarities, investigating designs at multiple scales can generate insight and the cross-fertilization of ideas. Some of the earliest mechanical m...

We are constructing a knowledge-based system that will aid astronauts in the performance of vestibular experiments in two ways: it will provide real-time monitoring and control of signals and it will optimize the quality of the data obtained, by helping the mission specialists and payload specialists make decisions that are normally the province so...

Thesis (Ph. D.) State University of New York, 1977. Typescript.