Novel Dual-Use Robotic Vectorization Platform for Abstraction Reasoning Interpretation, and Extrapolation in Hostile Environments Involving Occlusions and Noise

Autonomous on-orbit servicing, assembly, and manufacturing remains a daunting problem for many space-based use cases.  A system that could improve the speed and reliability of human-assisted and fully autonomous behavior in space would be highly desirable.  An extremely high degree of authority and situational awareness will be essential for both simple and more complex tasks and procedures. While the dexterity and speed of current robotic systems has improved dramatically over the last few years. In most cases, these systems are limited to telepresence under human supervision and pre-programmed activities in highly engineered environments.  Artificial intelligence and deep learning systems used for robotic analysis and pattern recognition have a significant limitation: there is no way to explain how their decisions are made. Conventional edge detection techniques result in too many false positives; this is because they are not cognizant of shape and have no awareness of the photometric characteristics of physical objects or topographical features. Vy has developed a platform to vectorize and make decisions about “big data” imagery in real world environments involving occlusions, low signal to noise, and variable lighting conditions. We call this platform Shape-Based Modeling Segmentation or SBMS.  It uses mathematical models called Bézier curves and decision trees to vectorize visual and hyperspectral imagery and saves all of information to an industry standard database called SQL. Vectorization turns related pixels into mathematical functions with the key differentiator being the ability to query and build complex models using Boolean rules and widely available tools like Python to fuse this data with existing platforms. It has the advantages of being fully auditable, requires no training, and provides a high degree of authority for quality assurance and situational awareness. In other than a highly engineered environments, the difficulty of autonomous robotic activity is finding an object with a camera or other sensor and converting it into a set of angles and distances to move, at any angle, at any resolution. SBMS is a platform that solves this problem by providing detailed instructions to any actuator or robotic system involving a series of auditable rules and procedures that can be accomplished with human supervision or fully autonomously.