Work Domain Analysis for the Interface Design of a Sonobuoy System

Transcription

1 PROCEEDINGS of the HUMAN FACTORS AND ERGONOMICS SOCIETY 51st ANNUAL MEETING Work Domain Analysis for the Interface Design of a Sonobuoy System Huei-Yen Chen, Catherine M. Burns Advanced Interface Design Lab, University of Waterloo Waterloo, Ontario, Canada Tab Lamoureux Humansystems Incorporated, Guelph, Ontario, Canada Modern sonar systems have greatly improved their sensor technology and processing techniques, but little effort has been put into display design for sonar data. The enormous amount of acoustic data presented by the traditional frequency versus time display can be overwhelming for a sonar operator to monitor and analyze. In addition to their normal sonar tasks, sonobuoy system operators manage the deployment of sonobuoys and ensure proper functioning of deployed sonobuoys. This paper describes a work domain analysis carried out as part of an interface design project targeting the sonobuoy system on board a maritime patrol aircraft. The domain of sonobuoy management and the domain of tactical situation awareness were modeled separately to address the two different aspects of the operator's work. Information requirements were drawn from the two models. Some of these requirements have significant implications for new design ideas that were previously overlooked in displays for sonobuoy systems. INTRODUCTION Sonar (sound navigation and ranging) technology has played an important role in underwater warfare since World War I. While sonar technology itself has improved immensely due to increased computational power and advances in sensors, little research has examined the presentation of sonar data. Presented with large amounts of complex, real-time data, sonar operators must distinguish signals of interest from background noise to detect, classify, localize and track known and potential contacts. Traditional sonograms of frequency versus time, or more widely referred to as waterfall displays, are still the major component of sonar displays (Waite, 2000). More recently, major improvements to sonar displays have largely come from the use of automated detection and tracking algorithms, which are now part of most commercialized systems (Kessel, R.T. and Hollett, R.D. 2006). Color coding of acoustic data was also introduced to enhance the visibility of certain signals (e.g. Zelter, D. and Lee, J., 1995). Displays containing geographical representations of sonar sensed contacts are also found in some commercial sonar simulations (e.g. ASWTT by ECA-Sindel of Italy, and CAE s STRIVE-SONAR) and military projects (e.g. the IMPACT project at Defence R&D Canada Atlantic). The Naval Undersea Warfare Center Div. Newport RI and the Fraunhofer Center for Research in Computer Graphics (2004) in the United States collaborated on a large scale interactive data visualization project for undersea warfare applications. They developed an innovative set of sonar display tools called EZ-grams, which help a sonar operator transform raw sensor data into useful information by allowing them to rapidly test hypotheses and search for confirming data. This paper outlines the requirements analysis for the interface of an air-borne sonobuoy system on board a maritime patrol aircraft (MPA). Sonobuoys are expendable sonar systems that can be dropped by aircraft or ships in an underwater warfare mission. Depending on their class, sonobuoys are capable of undertaking passive, mono-static active or multi-static active operations. Sonobuoys can be used either independently or fully integrated within joint operations conducted by allied or coalition forces. An operator for the sonobuoy system not only fulfils the sonar tasks described earlier, they are also responsible for managing and deploying the sonobuoys. The Ecological Interface Design (EID) framework proposed by Vicente and Rasmussen (1992) was adopted for this project. This framework is intended for complex, realtime, and dynamic systems where operators solve problems using prior knowledge and experience. For these reasons, EID is a good approach for designing an interface for sonobuoy management. We discuss the analysis, the work domain models, and the design requirements that resulted from this process. METHOD A work domain analysis (WDA) was carried out to provide a complete description of the system through a functional decomposition of the system into five levels of abstractions: al Purpose, Abstract, Generalized, Physical, and Physical Form. The primary sources of information for the analysis were literature research into sonobuoys, their capabilities, and principles of deployment, and field investigations with sonobuoy operators, and instructors. Two interview sessions were held with an acoustic operator from the Canadian Air Force and a sonar instructor from the Canadian Forces Naval Operations School (CFNOS)

2 PROCEEDINGS of the HUMAN FACTORS AND ERGONOMICS SOCIETY 51st ANNUAL MEETING at Halifax. Three group interviews were also held with nine officers from the navy, the air force, and the CFNOS. The group discussions were primarily on networked underwater warfare, but also allowed further understanding of sonar. We also communicated with the acoustic engineers responsible for the upcoming sonobuoy designs. Documentation regarding the sonobuoys and the towed arrays in development was reviewed, and manufacturer specification of the sonobuoys currently used by the Canadian Navy was also referenced. A literature search concerning the general practice of sonar in a military context was also conducted. System Description and Boundary Before a WDA is performed, it is important to consider the scope and purpose of the study and define the system accordingly. In this study, the objective of the analysis is to model the work environment for a sonar operator on board a MPA to support future interface design for the sonobuoy system. In this work environment, the operators handle distributed resources to address the need for tactical information over a defined period of time (the mission length). The operators' decisions and actions are bounded by military rules and practices, mission priorities, economic values, and environmental constraints. While the operators have control over their own assets (i.e. sonobuoys), they have little or no control over the natural environment and the actions of their known and potential contacts. Therefore we defined a loosely bounded system that contains the sonobuoys, contacts (which may come and go throughout the mission), and the natural environment in which the buoys operate. The meaning and purpose of a loosely bounded system was explained in Hajdukiewicz, J.R., Burns, C.M., Vicente, K. J. & Eggleston, R.G. (1999). Work Domain Models Results from interviews with subject matter experts identified two functional categories for our system operators: the management and deployment of the sonobuoys, and the monitoring and analysis of sensed data to build tactical pictures. Therefore, a two-part work domain model was constructed to reflect the two different emphases within the problem space. The first part modeled sonobuoy management, which is concerned with the physical allocation and proper functioning of sonobuoys. The second part modeled the domain of tactical situation awareness, a domain heavily structured by intentional constraints (characteristics of intentional systems as defined by Hajdukiewicz et al, 1999). However, it should be emphasized that the two domains are tightly coupled in achieving their purposes. Constraints from the social and natural environments apply to both domains, and the outcome or decision made in one domain supports the decision making in the other domain. The resulting models contained five levels of abstraction, following the guidelines proposed by Burns and Hajdukiewicz (2004). Three levels of decomposition were proposed for the domain of sonobuoys management: overall environment, sonobuoy network, and individual sonobuoy. The domain of tactical situation awareness was decomposed to the levels of overall environment, all contacts, and individual contact. The overall system objectives, which are to maximize chance of success in the mission and to meet naval values (rules, practices, and social norms), are reflected in the functional purposes of both domains. Figures 1 and 2 show the abstraction hierarchies constructed for the two domains, respectively. In the domain of sonobuoy management, the al Purposes are to maximize mission success by maximizing sonar signal detection ability in a given area. As well, other purposes in this domain are to minimize the number of sonobuoys deployed and to meet naval values. At the level of Abstract, natural laws of mass and energy were included to provide a basis for environmental elements. Flows of resources, authority, and economic values were modeled, as well as principles of military tactics and intelligence. Operators must effectively balance the need to achieve mission success and minimize sonobuoy deployment. The Generalized level modeled the air and water processes, processes for deploying and configuring sonobuoys and the networks they formed, and finally procedures for confirming and operating within limits. The Physical level mainly described the characteristics and capabilities of the environment and the sonobuoys, and documentation and guidelines for the naval practices involved. At the lowest level of work domain model, Physical Form, the model accounted for the physical attributes of the environmental characteristics, sonobuoys, and the specifics of naval guidelines and other documentation. The domain of tactical situation awareness contains many components that are either identical to or similar to components from the first domain. However, in this case the model is focused on locating and identifying contacts. The al Purposes in this domain are to maximize the number of known contacts and maximize the completeness and accuracy of the tactical picture compiled. At the same time, it is just as important to minimize time required to collect and understand contact information. As in the previous model, operations within this domain are required to meet naval values. At the Abstract level, natural laws, military and social values, and rules from the other domain still apply; principles of underwater sound and principles of geometry are introduced to govern the use and interpretations of acoustic data. Processes included at the Generalized level are signal detection, localization, and tracking; the aggregation and verification of environmental and tactical information; and confirming and operating within limits. The Physical level describes the factors that facilitate or limit how signals of some sources are sensed. This includes the characteristics and capabilities (or limitations) of contacts that expose them to the sonobuoy sensors; the environmental characteristics that affect the acoustic signal transmitting

3 PROCEEDINGS of the HUMAN FACTORS AND ERGONOMICS SOCIETY 51st ANNUAL MEETING Figure 1: Abstraction hierarchy of the domain of sonobuoy management Figure2: Abstraction hierarchy for the domain of tactical situation awareness

4 PROCEEDINGS of the HUMAN FACTORS AND ERGONOMICS SOCIETY 51st ANNUAL MEETING capability; and the resources for monitoring and analyzing acoustic data. Naval guidelines are also included to specify limits of the operation. At the Physical Form level, the environmental traits are identified; actual data, time, status, classification or any other attributes of contacts are described; and the location of the operation boundary is also stated. DESIGN IMPLICATIONS The work domain models enabled constraints, relationships between components, and means-end relationships across abstraction levels to be drawn within each domain. Tables 1 and 2 present some of the information requirements extracted from the sonobuoy management model and the tactical situation awareness model, respectively. The requirements shown in italics are known available information sensed by the current sonobuoy system. Table 1: Information requirements: the sonobuoy management model. Abstraction Information Requirement Level al Number of sonobuoys deployed. Range of Purpose possible sonar coverage by deployed sonobuoys. Chance of signal detection (density of sonobuoys in the area, high level assessment of environmental Abstract Generalized Physical Physical Form conditions). Water mass and energy levels. Sonobuoys storage level and rate of deployment for each class of sonobuoys. Level of adherence to policy and procedures. Propagation level of acoustic signals, air movement processes, range and accuracy of deployment, aircraft movement, selected settings of each sonobuoy (frequency channel, depth). Pattern planned for sonobuoys deployment. Actual pattern of deployed sonobuoys. Wind speed/direction, air temperature, and atmospheric pressure. Water temperature and pressure profiles. Direction and speed of current. Capabilities of deployment pattern. Anticipated battery life of sonobuoys. Sea bottom composition and contour, water temperature, salinity, and pressure, at various depths. Locations (bearings) of sonobuoys. Remaining battery life, radio signal strength, shape, size, visible (color) marking, cost, operational status, activation time, class and type of individual sonobuoys. Location of Rules of Engagement boundary. Table 2: Information requirements: the tactical situation awareness model. Abstraction Information Requirement Level al Time to establish a contact. Classification Purpose level of contacts successfully established. Number of contacts established during the mission. Percentage correction in determining a contact s location and classification. Completeness of tactical Abstract Generalized Physical Physical Form picture. Water mass and energy levels. Propagation level of underwater sound. Level of adherence to policy and procedures. Probable regions of contact location. Assessment of contact threat. Sound speed profiles. Water movement processes. Documented acoustic signatures of known and expected contacts. Location, signal pattern, and strength of known and potential contacts. Displacement of contacts being tracked. Water temperature and pressure profiles. Direction and speed of current. Constraints and difficulties in signal detection posed by ocean floor characteristics. Strength of signals emitted by contacts. Possible actions permitted within operational limits. Sea bottom composition and contour, water temperature, salinity, and pressure, at various depths. Acoustic signals sensed by sonobuoys. Current and historical results (detection, localization, tracking, and classification details of contacts). Location of Rules of Engagement boundary. As seen from the tables, attributes from the physical function and physical levels are more likely to be readily available. Tactical sonobuoys provide frequency vs. time information of the sonar data, physical locations of the sonobuoys, and settings pre-selected by the operator. Basic environmental conditions, such as water temperature and pressure, are sensed by a special type of sonobuoys, the bathythermal sonobuoy. The other information is not directly provided by the system and demands further data processing, often in the form of a paper-based calculation or a mental assessment by the operator. At the functional purpose level, information requirements reflect potential measures of the system in fulfilling its objectives. However, an accurate assessment of the correctness of the tactical picture compiled is almost impossible to obtain during a realistic mission. Information requirements and design criteria also arise from the interdependency of the two domains. On-going decisions regarding sonobuoy deployment are based on the tactical picture constructed. The need to assess the tactical

5 PROCEEDINGS of the HUMAN FACTORS AND ERGONOMICS SOCIETY 51st ANNUAL MEETING value of data sensed by a particular sonobuoy, considering its capability and location, stands out as a multivariate, crossdomain relationship to be accounted for in the interface design. CONCLUSION The work domain analysis presented for the sonobuoy system yielded significant information requirements for the design of displays for deploying sonobuoys and using sonobuoy data to identify contacts. Several information requirements had not been previously identified and are leading to new display concepts for a sonobuoy system. For example, the requirement of including sonobuoy inventory and usage information has not been addressed in the past designs of sonobuoy systems. A visualized inventory count would allow the operator to quickly determine their available resources. Similar display concepts can also be used to satisfy the operator s need to calculate limits in resources for proposed changes in sonobuoy deployment rate. Environmental conditions were also identified as valuable information for supporting a sonobuoy system operator s decision making. While water temperatures, pressures and salinity are sensed directly, their impact on underwater sound propagation is largely determined by a holistic view. Sound speed profiles, for example, are heavily utilized by sonar operators. In the advanced sonar systems used today, plots of sound speed profiles may be generated automatically; however, the perceived significance of the plots, such as potential blind zones of detection and the impact of variability over time, is still dependent on an operator s experience and training in underwater acoustics and warfare. Displays revealing water and ocean floor characteristics would provide extra support in analyzing sonar data. We also acknowledged a need for an operator to be able to maintain an overall tactical awareness while attending to data from a single sonobuoy. A resulting design criterion is to provide means to appropriately alert the operator. This could be accomplished using gaze-direct or other transitional tools for the effective monitoring of data arriving from multiple sources. In the future stages of this project we will develop displays based on the information requirements extracted from the work domain analysis. We will evaluate these displays to determine their ability to help operators to deploy sonobuoys efficiently and effectively, and to create an accurate and complete tactical picture for tracking and identifying contacts in an underwater warfare mission. participated in interviews and provided access to their work environment, and Jacquelyn M. Crebolder for making the interviews possible. The authors would also like to thank those who gave feedback in the writing of this paper. REFERENCES Burns, C.M. and Hajdukiewicz, J.R. (2004) Ecological Interface Design. Boca Raton, FL: CRC Press. Hajdukiewicz, J.R., Burns, C.M.., Vicente, K. J. & Eggleston,R.G. (1999) Work Domain Analysis for Intentional Systems. Proceedings of the 43 rd Annual Meeting of the Human Factors and Ergonomics Society, Kessel, R.T. and Hollett, R.D. (2006). Underwater Intruder Detection Sonar for Harbour Protection: State of the Art Review and Implications. NATO Report NURC-PR Naval Undersea Warfare Center Div Newport RI (2004). Large Scale Interactive Data Visualization for Undersea Warfare Applications. NATO Report. RTO-MP-105. Vicente, K.J. and Rasmussen, J. (1992) Ecological Interface Design: Theoretical Foundations. IEEE Transactions on Systems, Man, and Cybernetics, 22, Zelter, D. and Lee, J. (1995). Research on Improved Sonar Displays: A Human/Machine Processing System. RLE Technical Report No. 595 ACKNOWLEDGEMENTS This research has been supported by the Industrial Postgraduate Scholarship of the Natural Sciences and Engineering Research Council of Canada (NSERC) and Humansystems Inc. in Guelph, Ontario. The authors would like to thank the members of the Canadian Navy who kindly