Organizational Unit:
Cognitive Engineering Center
Cognitive Engineering Center
Permanent Link
Research Organization Registry ID
Description
Previous Names
Parent Organization
Parent Organization
Organizational Unit
Includes Organization(s)
ArchiveSpace Name Record
Publication Search Results
Now showing
1 - 5 of 5
-
ItemQuantitative Assessment of Human Control on Landing Trajectory Design(Georgia Institute of Technology, 2010-12-02) Chua, Zarrin K.An increased thirst for scientific knowledge and a desire to advance humanity's presence in space prompts the need for improved technology to send crewed vehicles to places such as the Moon, Mars, and nearby passing asteroids. Landing at any of these locations will require vehicle capabilities greater than that previously used during the Apollo program or those applied in Low Earth Orbit. In particular, the vehicle and the on-board crew must be capable of executing precision landing in sub-optimal landing conditions during time-critical, high-stakes mission scenarios, such as Landing Point Designation (LPD) , or the critical phase of determining the vehicle's final touchdown point. Most proposed solutions involve automated control of landing vehicles, accepting no input from the on-board crew - effectively relegating them to payload. While this method is satisfactory for some missions, an automation-only approach during this critical mission phase may be placing the system at a disadvantage by neglecting the human capability of [what?]. Therefore, the landing system may result in a lack of dynamic flexibility to unexpected landing terrain or in-flight events. It is likely that executing LPD will require an ideal distribution of authority between the on-board crew and an automated landing system. However, this distribution is application-specific and not easily calculated. Current science does not provide enough detailed or explicit theories regarding allocation of automation, and the advantages provided by biological and digital pilots (either acting as the sole authoritarian or as a coordinated team) are difficult to describe in quantitative measures. Despite previous experience in piloting vehicles on the Moon, few cognitive models describing the decision-making process exist. The specialization of the pilot and the application pose significant practical challenges in regular observations in the target environment. The lack of quantitative knowledge results in predominantly qualitative design trade-offs during pre-mission planning. While qualitative analyses have proven to be useful to the mission designer, an understanding founded on quantitative metrics regarding the relationship between human control and mission design will provide the sufficient supplementary information necessary for overall success. In particular, increased knowledge of the impact of human control on landing trajectory design would allow for more efficient and thorough conceptual mission planning. This knowledge would allow visualization of the flight envelope possible for various degrees of human control and help establish conceptual estimations of critical mission parameters such as fuel consumption or task completion time. This report details an experiment undertaken to further understanding of the impact of moderate degrees of human control on landing trajectory design or vice versa during LPD. This report briefly summarizes current understanding and modeling of moderate control during LPD and similar applications, reviews previous and current efforts in implementing LPD, examines the pilot study to observe subjects in a simulated LPD task, and discusses the significance of findings from the pilot study.
-
ItemAnalysis of Human-System Interaction For Landing Point Redesignation(Georgia Institute of Technology, 2009-05-26) Chua, Zarrin K.Despite two decades of manned spaceflight development, the recent thrust for increased human exploration places significant demands on current technology. More information is needed in understanding how human control affects mission performance and most importantly, how to design support systems that aid in human-system collaboration. This information on the general human-system relationship is difficult to ascertain due to the limitations of human performance modeling and the breadth of human actions in a particular situation. However, cognitive performance can be modeled in limited, well-defined scenarios of human control and the resulting analysis on these models can provide preliminary information with regard to the human-system relationship. This investigation examines the critical case of lunar Landing Point Redesignation (LPR) as a case study to further knowledge of the human-system relationship and to improve the design of support systems to assist astronauts during this task. To achieve these objectives, both theoretical and experimental practices are used to develop a task execution time model and subsequently inform this model with observations of simulated astronaut behavior. The experimental results have established several major conclusions. First, the method of LPR task execution is not necessarily linear, with tasks performed in parallel or neglected entirely. Second, the time to complete the LPR task and the overall accuracy of the landing site is generally robust to environmental and scenario factors such as number of points of interest, number of identifiable terrain markers, and terrain expectancy. Lastly, the examination of the overall tradespace between the three main criteria of fuel consumption, proximity to points of interest, and safety when comparing human and analogous automated behavior illustrates that humans outperform automation in missions where safety and nearness to points of interest are the main objectives, but perform poorly when fuel is the most critical measure of performance. Improvements to the fidelity of the model can be made by transgressing from a deterministic to probablistic model and incorporating such a model into a six degree-of-freedom trajectory simulator. This paper briefly summarizes recent technological developments for manned spaceflight, reviews previous and current efforts in implementing LPR, examines the experimental setup necessary to test the LPR task modeling, discusses the significance of findings from the experiment, and also comments on the extensibility of the LPR task and experiment results to human Mars spaceflight.
-
ItemContextual Inquiry of a 50 Aircraft Regional Airline Systems Operation Center(Georgia Institute of Technology, 2007-02) Feigh, Karen M.A contextual inquiry was conducted at the Systems Operations Control (SOC) of a regional airline with approximately 50 aircraft from the 8th-11th of November 2006. A total of 35 hours of direct observation were conducted with various members of the SOC Staff including the System Operations Control Shift Manager (SOCSM), the System Customer Service Manager (SCSM), the Dispatchers, and the Line Maintenance Planners (LMP). During the inquiry a wide variety of situations occurred: unscheduled maintenance delays, estimated ready time slips, a lightning strike, aircraft damage from a ground vehicle, a system-wide gate printer outage during a departure push, ATC delays, internet and subsequent ACARS outage, an unruly passenger disruption and turn back, and a sick dispatcher. The vast majority of these situations were handled as if they were no different from routine operations; however, there were moments when the SOC personnel were fully involved in the situation, and other minor tasks were being ignored or transferred to other personnel. The majority of high impact problems faced by the the airline’s SOC on a daily basis came from unscheduled maintenance or IT glitches. Unlike other airlines, ATC restrictions are not often an issue for this airline, although station curfews in southern California do place an additional constraint on the schedule recovery process. Similarly, weather was also only a minor issue during the contextual interview. Beyond the inevitable weather and maintenance interruptions, the majority of problems stemmed from software tools which limited the efficiency of the SOC personnel, and from procedures that required the SOCSM to do certain steps multiple times. For example, in order to keep the non-SOC personnel informed about the state of the airline, the SOCSM is required to run reports after each routing change and paste them into both email and the shift log. Additionally, the SOCSM is required to manually enter flight data to create new flights or to maintain existing ones. Similarly, the SOCSM is also required to manually enter and maintain maintenance segments for aircraft. The solution to these problems includes making better use of the current software’s functionality, investigating the actual information needs of the routing change recipient list, and incorporating additional automation to automatically create routing change reports and shift logs. The current software includes a capability to create new flights or maintenance segments using a correctly formatted text file. Using this capability would save much time in manual entry and minimize the number of typographical errors. Additional software should also be created to transition the incident reporting system and the shift log to an electronic database to facilitate data analysis. The SOCSM is currently responsible for posting any routing changes to a preset list via email. The actual information needs of these recipients should be reviewed to determine how frequently this information is actually required and whether or not a more scheduled reporting of all routing changes during a given time period might be adequate. Depending on the outcome, it might be possible to consolidate reports to once or twice a shift. Regardless, additional software should be created to automate the reporting process.
-
ItemContextual Inquiry of a Major US Airline Systems Operation Center(Georgia Institute of Technology, 2007-01) Feigh, Karen M.A contextual inquiry was conducted at the airline’s Systems Operations Control (SOC) from the 13-15th of November 2006. A total of 26 hours of direct observation were conducted with various members of the SOC Staff including several of the Operations Coordinators, the ATC Coordinators, and the Operations Manager. During the inquiry a wide variety of situations occurred: unscheduled maintenance delays, estimated ready time slips, multiple hub ground delay programs, severely reduced arrival rates due to cross-directional winds, ground delay program revisions, and diversions of international flights. The vast majority of these situations were handled as if they were no different from routine operations; however, there were moments when the key SOC personnel were fully involved in the situation and the normal coordination and collaboration between the ATCCs, OCs, MOC and crew coordinators reverted to top down command and control. Thus the workload is not evenly distributed across all SOC personnel because of the geographic distribution of responsibilities. In addition to these observations this inquiry identified three issues with specific design implications, all centered around the OC’s work practices: overly involved coordination sessions with MOC, lack of control of printer output, and the use of schedule printouts as a primary source of solution information. All three of these issues lead to inefficiencies in the SOC operation, despite which, however, the SOC in general and the OCs in particular are able to remain effective. This report suggests that the OCs could become more efficient by shedding some of their printer maintenance tasks, extended MOC coordination sessions, and more effectively using software tools. In order to achieve this high level of effectiveness the SOC personnel actively adapt their roles and the balance of power depending on the level of operational disruption. With the addition of an MOC representative in the SOC or the availability of key maintenancerelated scheduling data, increased effectiveness may also be achievable under conditions of limited disruption. Changing the flow of messages from the printer to an on-screen system will help minimize the ‘busy’ work associated with maintaining the printer and keeping up with the printouts. Introducing new hardware and software tools to aid with the schedule sorting and filtering may also provide increased efficiency, especially for the more junior OCs.
-
ItemContextual Inquiry of a 100 Aircraft Regional Airline Systems Operation Center(Georgia Institute of Technology, 2006-09) Feigh, Karen M.A contextual inquiry was conducted at the Systems Operational Center (SOC) of a Regional Airline with approximately 100 aircraft from the 24-27th of July 2006. A total of 30 hours of direct observation were conducted with various members of the SOC Staff including the Director of Systems Operations (DSO), the Manager of Customer Operations (MCO) and the Line Maintenance Planner (LMP). During the inquiry a wide variety of situations occurred: unscheduled maintenance delays, estimated ready time slips, a bird strike, a disruptive passenger requiring a cabin lock-down, a declared emergency due to oil temperature, taxi delays, weather delays, and brake-cooling delays. The vast majority of these situations were handled as if they were no different from routine operations; however, there were moments when the SOC personnel were pushed to their professional limits and the introduction of any other, even minor, issue could have caused severe disruptions to the schedule. The majority of problems faced by the the airline’s SOC on a daily basis came from lack of resources (planes and flight crew) and from inclement weather. During the inquiry, between 4-12 planes ( 6-9% of the fleet) were consistently out for unscheduled maintenance. Additionally, one one day during the observations 241 flight crew who were scheduled to fly were unavailable. Unlike other airlines, ATC restrictions are not often an issue for this airline, although station curfews in southern California do place an additional constraint on the schedule recovery process. Beyond the resource shortages and the inevitable weather interruptions, the majority of problems stemmed from software tools which limited the effectiveness of the SOC personnel. For example, several of the major software tools depend on different databases with limited connectivity, creating discrepancies between systems and requiring information to be entered multiple times. Additionally, the VisOps tool, used a primary measure of airline schedule adherence, does not support the logging of problems/issues, solution generation through the use of either advanced sort and search features, optimization algorithms and solution sharing. To make best use of the software tools on hand, especially VisOps, larger computer monitors are needed. The resolution at which the software tools must be set for visibility limits their usefulness with 19 inch monitors. Finally, none of the staff interviewed could indicate to any consistent quantitative feedback regarding the relative merits of their decisions on overall system performance. Instead, they often faced inquires about specific decisions which may only make sense when viewed from the overall context of the situation. Appropriate feedback could be provided as summary statistics regarding number of fights canceled, average delay and daily operational costs, which could be generated and displayed to them automatically.