Context-Augmented Robotic Interaction Layer (CARIL) Phase II

Period of Performance: 06/08/2015 - 06/07/2017

$1.5MM

Phase 2 SBIR

Recipient Firm

CHI Systems, Inc.
2250 Hickory Road Suite 150
Plymouth Meeting, PA 19462
Firm POC
Principal Investigator

Abstract

CHI Systems and the Institute for Human Machine Cognition have teamed to create a human-robot interaction system that leverages cognitive representations of shared context as a basis for a fundamentally new approach to human-robotic interaction. This approach centers on a framework for representing context, and for using context to enable robot adaptive decision-making and behavior. The framework is called CARIL (the Context-Augmented Robotic Interaction Layer). Context is an important part of human-human interaction. Unfortunately, context is often overlooked when designing robotic systems. The challenge is to translate high-level concepts, such as teamwork and collaboration, into specific requirements that can be implemented within control algorithms, interface elements, and behaviors. During Phase I, CHI Systems developed a proof-of-concept CARIL implementation and applied it to a notional simulated robot in a simple station model. This simulation demonstrated CARIL's feasibility by demonstrating how it gave the simulated robot a capability to reason about its context to avoid spatial interference with astronaut activities and tasks.