Traclabs (United States)

Changes GUI Fixed crashing on exiting OSG + ImGui applications: #1303 Compilers Tested Linux GCC 64-bit: 5.4.0, 7.3.0, 8.2.0 GCC 32-bit: 5.4.0 macOS AppleClang: 9.1.0, 10.0.0

Collision detection Added a feature of disabling body node pairs to BodyNodeCollisionFilter: #911 Kinematics/Dynamics Added setter and getter for WeldJointConstraint::mRelativeTransform: #910 Parsers Improved SkelParser to read alpha value: #914 Misc Changed not to use lambda function as an workaround for DART python binding: #916

A critical resource for meeting the challenges of human exploration beyond near Earth orbit is greater reliance on spacecraft automation. If human-automation integration is not appropriately designed, however, increased automation may contribute to rather than alleviate these challenges. TRACLabs and San Jose State University are investigating the hypothesis that selecting units of work for automation based on procedures written for humans will improve human-automation designs. We conducted an exploratory study with human participants using TRACLabs' PRIDE software for authoring and executing electronic procedures to perform simulated spacecraft operations. Our research investigates manual, partially automated, and fully automated procedure execution, and support for these multiple modes. We report the results of participant use of automation functions in this paper.

This paper examines 29 papers that have proposed or applied metrics for human-robot interaction. The 42 metrics are categorized as to the object being directly measured: the human (7), the robot (6), or the system (29). Systems metrics are further subdivided into productivity, efficiency, reliability, safety, and coactivity. While 42 seems to be a large set, many metrics do not have a functional, or generalizable, mechanism for measuring that feature. In practice, metrics for system interactions are often inferred through observations of the robot or the human, introducing noise and error in analysis. The metrics do not completely capture the impact of autonomy on HRI as they typically focus on the agents, not the capabilities. As a result the current metrics are not helpful for determining what autonomous capabilities and interactions are appropriate for what tasks.

Automated systems, ranging from robots to intelligent planning aids, are increasingly part of the cognitive landscape. A pressing question facing the human factors community is how to design the automation to foster effective collaboration between the human and automated agents. This panel brings together leading researchers actively engaged in design of decision-support systems that involve some element of ‘cognitive work’ automation. They represent a variety of domains and a variety of approaches to design of collaborative automation. Panel members will discuss unique challenges of their respective domains, including differences in temporal rhythm, and how these have shaped their perspective on ‘collaborative aiding’.

We previously developed a core representation for describing the information a human needs to perform a task and the information provided by a user interface. This representation is highly abstract and is based on information theoretic properties, thus it can be applied to a wide variety of work domains and information and display types. Since information need and information conveyed are described in the same numerical scales, it is straightforward to compute a degree of match between them. In prior work, we used this capability to dynamically and automatically reconfigure cockpit displays for military cockpits. In recent work, however, we adapted this approach to the task of evaluating and critiquing display format designs to support procedure execution in the context of NASA's space operations. The representation and reasoning approach generalizes well to describing information types in procedural domains. The resulting tool can be used to (a) analyze a proposed display format for a given task, (b) propose a format for a given task, (c) project how changes to a procedure will affect the suitability of a previous format, and (d) project how changes to a format will improve or reduce its suitability for a given procedure.

We previously developed a core representation for describing the information a human needs to perform a task and the information provided by a user interface. This representation is highly abstract and is based on information theoretic properties, thus it can be applied to a wide variety of work domains and information and display types. Since information need and information conveyed are described in the same numerical scales, it is straightforward to compute a degree of match between them. In prior work, we used this capability to dynamically and automatically reconfigure cockpit displays for military cockpits. In recent work, however, we adapted this approach to the task of evaluating and critiquing display format designs to support procedure execution in the context of NASA's space operations. The representation and reasoning approach generalizes well to describing information types in procedural domains. The resulting tool can be used to (a) analyze a proposed display format for a given task, (b) propose a format for a given task, (c) project how changes to a procedure will affect the suitability of a previous format, and (d) project how changes to a format will improve or reduce its suitability for a given procedure.

Technical advances since Apollo make it possible to perform robotic reconnaissance to gain a better understanding of lunar sites prior to human exploration. NASA is conducting analog field tests to investigate these operations concepts with advanced robots and simulated flight operations. We have developed robot performance monitoring software for use during robotic reconnaissance operations. We measure robot performance by monitoring robot data in real-time and computing robot performance metrics from that data. Metrics are computed for two regimes of flight operations -- remote supervision of autonomous robot operations and debrief support after a flight operations shift. In this paper we describe our performance monitoring software, define the metrics we compute, discuss how these metrics are used in flight operations, and summarize results from recent field tests.

We describe progress in developing a software tool to help create, communicate, and refine concept of operations (ConOps) information. Information found in a ConOps document is central to our tool. ConOps documents are written in the end user's language, describing how the proposed system should support the users' tasks. While there are some helpful documents prescribing ConOps format, there is little direct assistance to express the concept of operations. We describe a tool to help author this information, integrate it with storyboard sketches, refine it through interaction with other stakeholders, and finally export the information in a form directly usable by developers. We have developed a demonstration prototype that performs the core functions of ConOps definition. We discuss the results of this prototyping effort and plans for expanding the tool functions. We plan to refine the tool through experience with ongoing projects so that it supports system concept development.