Carleton University Simulator Project (CUSP)

M.J.D. Hayes1, R.G. Langlois1, T.W. Pearce2, C.-L. Tan1, J.A. Gaydos1

1 Department of Mechanical & Aerospace Engineering, Carleton University, 1125 Colonel By Drive, Ottawa, ON, K1S 5B6, Canada [email protected], [email protected] [email protected], [email protected]

2Department of Systems and Computer Engineering, Carleton University, 1125 Colonel By Drive, Ottawa, ON, K1S 5B6, Canada [email protected]

This paper presents novel aspects of CUSP, an industrially relevant 4th year design project, and its pedagogi- cal paradigm in the framework of Capstone Design Projects in the Department of Mechanical and Aerospace Engineering at Carleton University. There are currently six projects involving 28 faculty members, 6 graduate students, 200 4th year students, and a signiﬁcant number of 3rd year student volunteers. CUSP will be presented as a case study. This project is by nature multidisciplinary and includes participation of the Depart- ment of Systems and Computer Engineering, the Centre for Applied Cognitive Research, and the Eric Sprott School of Business. Given its industrial relevance, proposals have been made for industrial sponsorship of graduate students and post-doctoral fellows to lead basic research and development aspects, thereby evolving the project into a vertically integrated research programme. A similar thrust is being made by the ﬁve other multidisciplinary projects.

1 INTRODUCTION veloped at Canadian universities. The most generally accepted idea was that modern engineer- In February 2003, the CSME and the Department ing design challenges require a multidisciplinary of Mechanical and Industrial Engineering at Con- paradigm, rather than an interdisciplinary team cordia University in Montreal´ co-hosted the Inter- approach. That is, in the paradigm of 4th year national Conference on the Future of Engineering capstone design projects it is not enough to en- Education (CSME-ICFEE 2003). The objective able interaction between different engineering dis- was to create an opportunity for frank discussion ciplines. What is required is the enablement of between engineers in academia and those in in- interaction between different disciplines. For ex- dustry. Topics ranged from issues facing women ample, a design team examining issues associated in engineering to a panel discussion on Interdisci- with human factors in the design of a human- plinary Engineering Programmes and another on machine interface must include not only mechan- Re-Engineering the Aerospace Curriculum. All ical, electrical, and systems engineers, but also re- of the sessions were productive, but the two panel quires input from psychologists, cognitive scien- discussions were particularly so. Moreover, the tists, industrial designers, physiologists, etc.. resulting conclusions and recommendations from The panel on re-engineering the aerospace cur- the two panel sessions meshed very nicely with riculum was chaired by Hany Moustapha from th the evolution of 4 year capstone design projects Pratt & Whitney Canada. The majority of panel in the Department of Mechanical and Aerospace members were from the aerospace corporations Engineering at Carleton University. in the Montreal´ area, including CAE, Bell Heli- Filippo Salustri, from the Department of Me- copter, and Bombardier. There was general agree- chanical, Aerospace, and Industrial Engineering ment that mechanical and aerospace engineering at Ryerson University, led the discussion on in- students graduate with sufﬁcient technical skills, terdisciplinary engineering programmes [2]. The but lack key soft skills. The outcome of the dis- panel members were all academics and the discus- cussion was a wish list of soft skills the aerospace sion focused on how such programmes can be de- industry members wanted the aerospace, and by

CSME 2004 Forum 1 extension the mechanical engineering, curriculum 2 EVOLUTION to impart to new graduates. This wish list was Carleton University began offering Masters and nicely summarized by Gerhard Serapins [3], Man- Doctoral degrees in aeronautical engineering in ager of Research and Development, Operations, the early 1960s. By the 1970s there were sev- CAE Inc.: eral faculty members who had spent some time in the United Kingdom and were well aware of “There is a need for students to expe- the design project that was a major part of the rience working in a virtual enterprise educational activities at The College of Aeronau- environment. Among other soft skills, tics, Cranfield, England (now Cranfield Univer- they need to experience a design project sity). The Cranfield project involved a design matrix: experience having to prioritise team of about 20 students, guided by several fac- among multiple supervisors and mul- ulty members. Over a period of a year or more tiple tasks; communicating in a large the team would typically develop, to quite a high multidisciplinary team; develop verbal level of detail, the design of an aircraft for some and written communication skills, but specified mission. also develop the capacity for unbiased In 1986 a proposal to offer an undergraduate listening.” degree program in aerospace engineering was being prepared for Carleton University’s Senate and For the last ten years the paradigm for the 4th there was a consensus that a team design project year capstone design projects in the Department similar to that at Cranfield should be included in of Mechanical and Aerospace Engineering at Car- the final year of the new program. The intent was leton University has evolved such that it satisfies to simulate insofar as possible the team-design en- the major requirements agreed upon by both panel vironment typically found in the aerospace indus- discussions [1]. That is, a compromise has been try. It was felt that such a team project would found that meets the academic requirements for not only provide students with substantial first- accreditation, while focusing on industrially rele- hand design experience but would also provide vant design issues. a vehicle for attaining other educational objectives, including making students aware of the im- While the projects are resource intensive both portance of collaborative effort, communications, in terms of funding, time, and space, the end re- documentation and configuration control and for sult is very well justified. Moreover, the return on giving them opportunities to improve their presen- the investment is irresistible. It includes good will tation and report-writing skills. The proposal was and cooperation from industry and government in- accepted by Carleton’s Senate and both an aircraft stitutions, research opportunity, potential graduate and a satellite design project, each with about 20 students, and graduating students well prepared to students, were implemented in 1991-92, the final make a strong contribution to any modern design year of the first Aerospace Engineering graduat- project. One very tangible benefit for our 4th year ing class. students that stems from the relationship between industry and university fostered by the capstone Representatives of industry were involved in design projects is a job. The final design reviews the projects from the beginning, both as lead en- are well attended by relevant industry represen- gineers in the project work itself and as evaluators tatives. Frequently students are invited to inter- in the formal design reviews at the end of each views, or even offered an entry level position dur- academic year, see [4], for example. The feed- ing the banquet following the design review. back from industry representatives was very pos- itive, so much so that in 2001-02 the same team In the next section we shall give a brief history th project format was adopted in the mechanical en- of the evolution of the 4 year capstone design gineering degree program. At present there are six project paradigm, present an overview of man- ongoing projects. agement and operational practices, and briefly describe each of the current six projects. The sub- 3 OVERVIEW sequent section will illustrate the concepts using CUSP as a case study, wherein we shall describe The current success and popularity of the capstone the objectives of CUSP in greater detail, discuss design projects largely can be attributed to three the management and organization of the matrix main factors: challenge, industrial relevance, and format, and outline the technical aspects. continuity. Experience has demonstrated that the

CSME 2004 Forum 2 more challenging the task faced by students, the At present, projects are proposed and selected greater the accomplishment that can be expected. by departmental faculty members. As industry To foster this, project managers and lead engi- fully appreciates the potential of these projects neers (faculty members and consultants) are find- for tackling challenging problems that can more ing that technical and management support rather easily be undertaken in a university environment than direction is an effective operating model for due to greater academic resources, greater cost ef- the projects. Each project is technically challeng- fectiveness, and reduced requirements for guaran- ing and generally based on ambitious end objec- teed success or immediate results, it is likely that tives. All the projects have multi-year time lines, projects will be selected and coordinated collabo- typically five years, carried out in yearly phases. ratively with the research and development activi- Projects are industrially relevant in two senses ties of industrial and government partners. - technically and administratively. The technical All projects include prototype fabrication relevance of projects facilitates obtaining finan- and testing of the entire design or major sub- cial support and interaction between students and assemblies. In the time frame of one academic engineers in industry. Currently there are more year moving from concept to prototype is an im- than 50 industrial participants supporting in cash, mense challenge for any design team. To achieve and in kind, the six projects. The combined in- this objective requires strict adherence to sched- dustrial and educational environment exposes stu- ules and budgets, but imparts in students an ap- dents to industrial project management, but with preciation that while “paper” is patient, moving to greater tolerance for mistakes along the learning physical reality through a prototype is not. The curve. The support provided by industry is partly requirement for prototype fabrication and testing due to the fact that these projects provide a pool means that design “timecompression” technolo- of highly qualified personnel who are already fa- gies must be developed and utilized, including miliar with technical and interpersonal aspects of concurrent engineering, CAD/CAM, virtual solid large projects, and with potential technical solu- modelling/visualisation, computer aided analysis, tions to difficult problems. Publicity with stu- numerical analyses, tradeoff and sensitivity stud- dents and the public at large, and an opportunity ies, and rapid prototyping. As importantly, stu- to contribute to engineering education are addi- dents must be innovative and efficient in their tional factors. Continuity results from both the communication, time management and interaction size of project teams and the multi-year nature of with industry and external participants. the projects. Three hour project general meetings are held A typical project now involves in excess of 25 every week to co-ordinate progress. Two formal students, five lead engineers, and a project man- design reviews are held - one each term. Industry ager, each contributing a minimum of 200 hours representatives familiar with the design objective (sometimes much more) to the project each year. participate in the final formal design review and In addition, one graduate student and numerous offer suggestions on design strategy, technical fea- 3rd year student volunteers participate in each of sibility and potential improvements. Projects in- the projects. This results in a project budget in ex- clude visits to industrial facilities of relevance to cess of 35,000 engineering hours over five years - the design objective. This collaboration and inter- representative of a large engineering project. The action with industry enhances the skills imparted large number of participants and long duration to students, enabling them to be more productive allow for efficient parallel developments as well when they embark on their careers following grad- as long serial developments such as design itera- uation. tion. While annual student turn-over necessitates To support these projects the Department more annual lead time than would otherwise be maintains Design Laboratories equipped with net- required, it provides many new perspectives and worked computers, the necessary CAD software, a continuous evaluation of previous design de- and specialized design and analysis software for cisions. One result is that the projects are cur- each project. In addition a wide range of proto- rently strong and are also evolving - quality is im- type fabrication, development and testing facili- proving, student involvement at the graduate level ties are available so that students can manage the and from programs external to Mechanical and transition of their designs from “paper” to physi- Aerospace Engineering is growing, industrial in- cal reality. terest and participation is increasing, and greater The spectrum the projects cover is broad financial resources are becoming available. and address virtually all aspects of mechanical

CSME 2004 Forum 3 and aerospace engineering. It is evident that ploiting, within financial constraints, new tech- industrially-motivated needs provide challenging nologies. The project is both complex and am- capstone design projects, and result in lasting ben- bitious and is therefore expected to span multiple efits both in terms of student development, as well years and correspondingly, multiple groups of stu- as potential employees and products. dents. The CUSP objectives can be categorized The six current design projects are: into both short- and long-term. Short-term objectives are defined based on a one year time line 1. Carleton University Simulator Project appropriate for a single class of students whereas (CUSP); long-term objectives are based on a five-year time 2. Air-launched Earth-observing Ground Infor- line. mation System (AEGIS); The long-term objectives of the project are to develop a complete and flexible simulation facil- 3. Zero-emission micro turbine engine; ity located at Carleton University including a va- 4. Autonomous Underground Mining Vehicle riety of mathematical models, a multi-functional (AUMV); motion platform, a general vision system, and 5. Formula SAE vehicle; a reconfigurable user interface all interoperating based on high-level architecture (HLA). This re- 6. Unmanned Air Vehicle (UAV). search and simulation facility will be used to support simulation education as well as specific 4 CUSP research objectives motivated by industry and CUSP is the newest of the capstone design government. Evidence suggests that such a fa- projects and was introduced in the 2002-2003 cility can eventually become economically self- academic year. The decision to develop a sim- supporting. ulation project was based on growing preva- lence of simulation throughout vehicle and process design cycles and for subsequent uses rang- ing from simulation-based acquisition through operator training. The concept of a simulator project has received strong support from the Canadian simulation community as it is projected that de- mand will exceed supply of recent graduates with the skill set required to integrate seamlessly in this industry over the next decade. Further, due to the strong support, significant opportunity exists for students to interact with counterparts in industry and government during their involvement with CUSP, and this interaction will further hone the important soft skills identified and highlighted by the panel discussions. Vehicle simulation in various forms has been applied to all types of vehicles including fixed- Figure 1: NASP conceptual design. and rotary-wing aircraft, surface and subsurface marine vehicles, on- and off-road ground vehi- The short-term objectives completed in Year cles, and rail vehicles as well as many process- 1 of the project (2002/2003) were to survey po- related environments such as power station op- tential applications, requirements, existing short- eration and air-traffic control. Simulator objec- comings, and possible configurations for simu- tives, structures, and specifications for each vehi- lator motion platforms that are appropriate for cle or process and application can be quite differ- a range of vehicle types, develop a preliminary ent. The overall technical objective of the CUSP design for a novel six-degree-of-freedom motion design project is to develop a set of requirements base (called NASP, for not a Stewart platform, for a multi-functional vehicle simulator and mo- and illustrated in Figure 1), and simultaneously tion platform that can be applied to a range of design and build an HLA compliant simulation vehicle types, and subsequently design the corre- demonstrator (called SiDFreD, for single degree- sponding multi-functional simulation facility ex- of-freedom demonstrator, and illustrated in Fig-

CSME 2004 Forum 4 the varied set of analytical skills and perspectives necessary to effectively advance a project of this scope. The organizational structure of the project is based closely on an industrial model where a faculty member serves as the Project Manager, while faculty members and graduate students serve as Lead Engineers responsible for leading functional teams. The students and lead engineers are organized into teams according to area of expertise (Integration-INT, Actuation-ACT, Kine- matics and Dynamics-DYN, Structures-STR, and Systems-SYS) and task-oriented groups (Business Development-BUS, SiDFreD Demonstrator-SID, NASP Platform-NSP, Safety Systems-SAF, Hu- man Factors-HMF, and Vehicle Simulation-VS). Each team includes approximately 6 students and each group is staffed as appropriate for the an- ticipated group activities. Table 1 illustrates the matrix concept. To facilitate communication and Figure 2: 4th year student Andrew Bruce in SiD- integration within the project, each team has an FreD testbed. appointed Integration Team liaison person and the group leaders are the members of the Integration Team. Lead engineers are also associated as tech- ure 2) having a single-degree-of-freedom based nical advisors to the groups. Using this matrix on a road vehicle system that includes a mathe- structure, a functional and industry-relevant man- matical model, user input, actuation, and graphi- agement structure is created. cal display. While initially a simple system, this The project has a weekly general staff meet- model establishes an approach that can be applied ing attended by all participants. These meetings for the implementation of the full motion simula- provide an opportunity to discuss technical as- tor. pects and progress of the overall project, listen to The short-term objectives for Year 2 of the presentations by invited speakers relevant to the project include developing SiDFreD into an effec- project, and present data and results that are of tive multi-functional simulator having three de- common interest to project members. The for- grees of freedom; refining and expanding the mat of weekly staff meetings includes opening re- NASP platform and facility into a practical multi- marks by the project manager and lead engineers, functional design; and strengthening external ties status reports from the project teams and groups, through focussed research and effective commu- general discussion, and individual group and/or nication relevant to the simulation industry. team meetings. In the event of guest speakers or facility tours, the format is altered as necessary. In addition to the common meeting, lead 5 ORGANIZATION engineers generally meet with team members at The current CUSP design team includes five fac- another convenient time as well. Student groups ulty members, one graduate student, and twenty- meet on their own as necessary. eight final year undergraduate students from pro- With the number of project members involved grams in Aerospace Engineering, Mechanical En- and the number of tasks undertaken concurrently gineering, Systems and Computer Engineering, and the many associated interdependencies, com- Computer Science, Buiness, and Psychology at munication between individuals, teams, groups, Carleton. In addition, the possibility exists for un- and the project management are vitally impor- dergraduate student volunteers and graduate stu- tant. In addition to verbal communication dur- dents to become involved with the project. In ing the weekly meetings, project information is the past academic year, five third-year engineer- documented and transferred using a combination ing student volunteers were directly involved with of brief technical memos to convey information CUSP. Note that the multi-disciplinary team offers to other team members, design reports to docu-

CSME 2004 Forum 5 Table 1: CUSP organizational structure

group \ team INT ACT DYN STR SYS Business student 1 student 2 ··· SIDFreD student 3 student 4 ··· . . . NASP . . .. Safety Human Factors Vehicle Simulation ment signiﬁcant accomplishments, and a compre- 5.1.2 SIDFreD hensive ﬁnal report summarizing the contribution to the project by entire classes of students. The The several integrated degree-of-freedom demon- objective is to have concise effective communi- strator (SIDFreD) Group is responsible for all as- cation. A project web site is used as a central pects of designing, integrating, and testing the means of communication and serves as a reposi- SIDFreD technology demonstrator. The range of tory for all ‘published’ information such that it is tasks includes design and manufacture, mechani- available to all project members at all times. The cal maintenance, software development and test- web site has both public and private components ing, and operation. Further, since SIDFreD is in- as the public portion of the web site is used to keep tended in part to be a technology demonstrator for project sponsors and potential industrial receptors the NASP design, the SIDFreD Group is respon- of the developed technology aware of progress to sible for customizing SIDFreD as appropriate for date. performing new technology evaluation.

5.1 Groups 5.1.3 NASP The task-oriented groups involved with CUSP are formed and staffed consistently with current The acronym NASP stands for not a Stewart Plat- project objectives and can vary from year to year. form, which reflects the strong emphasis on inno- As an example, the groups appropriate for Year vation within the projects. The NASP Group is 2 of CUSP are reflected in Table 1. Functions of responsible for developing concepts and advanc- individual groups are described briefly below. ing the design of the conceptual NASP platform. Unlike SIDFreD, that involves physical hardware, the NASP Group is focussed on an initially paper design that will necessarily require the collaboration of industry and government partners for its 5.1.1 Business Development ultimate fabrication and evaluation.

The two primary functions of the Business Devel- opment Group are sponsorship and external com- 5.1.4 Safety Systems munication. The funding for CUSP relies heav- ily on sponsorship provided from various sources The Safety Systems Group ensures that safety- within Carleton University, Industrial and Gov- related issues are addressed completely within the ernment contributions, discounts offered by sup- project. Activities include ensuring that CUSP pliers, and in-kind support provided by indus- meets or exceeds departmental, university, and try, government, and individuals. In this regard, provincial health and safety guidelines and en- the business group identiﬁes potential support sures safety aspects are included in both the me- and prepares the appropriate sponsorship pack- chanical and software aspects of both the SID- ages and applications. In the second capacity, the FreD facility and the NASP design. Safety is business group prepares project brochures, devel- highlighted throughout CUSP and innovative ap- ops and maintains the project web site, and inter- proaches for improving simulator safety systems faces with the media. are investigated.

CSME 2004 Forum 6 5.1.5 Human Factors managers. The conceptual design of NASP requires signiﬁcant research of the state of the art The Human Factors Group is focussed on two pri- of simulators while the proof-of-concept demon- mary issues. First, the cognitive science aspects strator requires aspects of the simulator to move of simulation are addressed in determining lim- off the drawing board and become a working pro- its of human perception such that they can be in- totype. INT must oversee this transformation. corporated into the platform washout algorithms thereby ensuring high ﬁdelity of the developed systems. Second, ergonomics aspects of the simulators are addressed by this group. 5.2.2 Actuation (ACT)

5.1.6 Vehicle Simulation Initial effort in the Actuation (ACT) team was The Vehicle Simulation Group is responsible for directed at understanding various techniques to the mathematical modelling and simulation of the achieve controlled motion including electromag- vehicle systems that will be simulated using the netic, ball screw, linear belt drives and related SIDFreD and NASP platforms. As the overall ob- motors. Research included a physical description jective is to develop reconfigurable platforms, spe- of the method of operation, its technical limita- cial attention is required to ensure that the ranges tions, its cost and an estimate of the time to manu- of motion of the platforms are appropriate for a facture not-in-house components. Subsequent ef- range of vehicle types. This activity of simulat- forts were directed at understanding the relations ing various vehicles is performed by the vehicle between the structural constraints (placement on simulation group. Further, the developed vehicle platform, force, torque), the envelope of motions models are integrated into the SIDFreD demon- that would be attempted (kinematics, velocities, strator and provide the core mathematical mod- accelerations) and the duel issues of human phys- els and simulation software driving the developed iology (what motions feel real) and cognitive per- simulators. ception (what motions appear real). Research on these issues and their inter-relationships are ongo- 5.2 Teams ing through group discussions, etc. Preliminary results permitted ACT to initiate a process to im- As mentioned earlier, the students and lead engi- plement a second degree of freedom on the exist- neers are organized into teams according to area ing SIDFreD platform. of expertise. Each team includes approximately 6 students and each student is cross appointed to The process to implement a second degree of one group so that each team has a representative freedom required additional team interaction with of each group, and vice versa. The scope and tech- the Systems team (SYS) to insure a proper se- nical objectives of the five teams are summarized lection of actuator PC control card and compat- below. ibility with the system architecture. Thirty in- dustries involved with actuator design and manufacture were identified and canvassed for product 5.2.1 Integration (INT) information including a product’s operating char- The focus of the integration team is primarily on acteristics, reliability, robustness, expansion flex- project management. Tasks include establishing ibility, cost, product support, ease of implemen- a project work breakdown, and critical time lines tation and integration with power supply, control for deliverable deadlines to ensure all tasks will card and a software interface. A point student be completed on time; coordinating the efforts of was identified and placed in charge of coordinat- all teams and groups; overseeing preparation of ing and cataloging the team’s efforts and accom- documentation and presentation templates. Inte- plishments during this phase of the effort. A sig- gration team members are required to be familiar nificant challenge was to optimize the informa- with all aspects of the project. tion obtained and the industrial relationships over Integration team members get their share of a relatively short period of time to achieve a good experience with technical aspects of the design fit with the aims of the CUSP project. The next through their cross appointment to one of the phase of the project will include extensive test- groups. Moreover, the Integration team members ing of the software controlled actuator to obtain must understand, at least superficially, all techni- load-frame information on the load-displacement, cal aspects of the project in order to be effective load-velocity and load-acceleration behaviour.

CSME 2004 Forum 7 5.2.3 Kinematics and Dynamics (DYN) the DYN team in collaboration with industry and the research community. The primary focus of the kinematics and dynamics team is on model development. These models are essential for platform analysis and con- 5.2.4 Structures (STR) trol. Because the control system requires position, In the first week of the Fall semester, the Struc- force and torque output in real time, the math- tures team is assigned the task of drafting sev- ematical models must be computationally opti- eral designs of the motion platform that can po- mized. The equations of motion for the 3 DOF tentially yield the range and the degrees of free- SIDFreD are reasonably straightforward, and em- dom required. The brainstorming of ideas typi- ploy the Newton-Euler formulation. But DYN is cally involve students from other teams as well. also responsible for washout algorithm develop- When accomplished, these ideas are presented at ment and implementation in the appropriate HLA the weekly general meeting of all the members of Federate. the project. The advantages and disadvantages of The kinematics and dynamics of 6 DOF mo- the different conceptual designs are discussed and tion platforms are non-trivial, and are still a hot one of them chosen for the detailed design for the topic in the research community. The DYN team year. will have to carefully examine traditional Carte- Students in the Structures team are then as- sian techniques as well as more sophisticated signed separate tasks of determining the final de- methods (kinematic mapping, Study’s soma, dual- tailed design of the different structural parts of quaternions, etc) to develop the kinematic model. the motion platform. This involves compiling the For the dynamics the Newton-Euler, Lagrange, list of components and carrying out stress analysis Kane, and possibly other formulations will be ex- to obtain the final physical dimensions to ensure amined. The bottom line shall be computational their structural integrity. Where necessary, this efficiency. may involve using commercial finite element soft- Human factors and Washout represent a key ware. Final design working drawings of the com- point of intersection between the fields of hu- ponents and the final assembly are then created man physiology, psychology, cognitive science, and documented. The design and manufacture of and engineering. The implementation of a con- SIDFreD follows a similar process as described vincing motion simulator starts by modelling how above, with a tighter deadline. This is feasible as the proprioceptive, visual, and vestibular recep- the demonstrator must be completed before the fi- tors respond to force and motion cues encoun- nal design review in March. tered during actual operation of the vehicle be- Once the design drawings of the components ing simulated. The training simulator must pro- and the assembly have been approved by the lead vide cues that the human receptors interpret as engineer of the team, the students then proceed real. Moreover, the simulator must not provide with sourcing and purchase of the materials re- negative training. These motion cues must be pro- quired, followed by the manufacture in the de- vided by a combination of environmental prompts partmental machine shop of the necessary parts and movement of the simulator platform within its that they have designed. Throughout this pro- operational workspace envelope. They must be cess, there is constant communication between the imparted such that the perception of the motion members of the Structures team and those of the and associated forces are sustained while the plat- Actuation and Dynamics teams in particular to en- form returns, with movement below the percep- sure that all their engineering and safety concerns tion threshold of the operator, to a neutral kine- are duly considered in the structural design. matic configuration to await the next command input. This is accomplished by a motion cueing 5.2.5 Systems (SYS) algorithm, or washout filter. To implement a washout filter the reachable The Systems Team (SYS) is responsible for the and orientable workspace limits of the motion computing infrastructure associated with CUSP, platform must be known. Workspace characteri- and the Lead Engineer for the team (Dr. Trevor zation of 6 DOF spatial parallel platforms is also a Pearce) was recruited from the Department of hot research topic among the kinematics commu- Systems and Computer Engineering. The team’s nity. Hence, there are no off the shelf solutions. responsibilities have expanded over the past two The characterization will have to be developed by years, and SYS members are active in the com-

CSME 2004 Forum 8 plementary groups that deal with issues that span at Carleton University, citing CUSP as a specific the project’s team structure. Membership in the example. The projects are broad in scope and team is open to any interested student; however, multidisciplinary in nature. While they are re- the computing focus has resulted in the team being source intensive, experience has shown that, when dominated by students from Computer Systems carefully managed, the results are well worth the Engineering, Software Engineering, and Commu- investment. A strong working relationship can nications Engineering programs. With the current be formed between industry, government, and SYS team representing about one quarter of the university research and development programmes total number of CUSP students, it represents a sig- which can lead to opportunities for all involved. In nificant infusion of interdisciplinary participation. the best case this means jobs for students; highly The scope of the SYS team has evolved with qualified personnel for industry; funding and col- the project. In the first year, the team focused on a laborative research opportunities for faculty. computing architecture suitable for the short and Much work remains to be done. To convince long-term goals of CUSP. A central goal was to other departments in the Faculty of Engineering include the High Level Architecture (HLA) for and Design, and in the university at large, to com- simulation interoperability [5]. The HLA pro- mit resources, faculty, and students requires care- vides a framework for component-oriented dis- ful planning. All participants require some tangi- tributed simulation, and was targeted as the under- ble return on the investment. This clearly enlarges lying infrastructure to enable the run-time com- the scope of the virtual enterprise analogy. While puting associated with CUSP platforms. Over the we must be committed to maintaining the princi- past two years, the team’s scope has expanded ples of collegiality, and be careful to not make de- to include various sensors, development environ- cisions that could erode our academic and philo- ments, and business planning. SYS members also sophical independence, we must strive to be rele- participate in broader cross-team activities asso- vant. We all think; may as well think big. ciated with safety, human factors of display technology, psychology, washout algorithms, manu- BIBLIOGRAPHY facturing, procurement, assembly, system integra- [1] Hayes, M.J.D., Langlois, R.G., Kind, R.J., tion and project management. “Mechanical and Aerospace Engineering The use of the HLA is not entirely without Capstone Design Projects at Carleton Uni- drawbacks. The HLA has a comprehensive set versity”, Proceedings of the International of services designed to support a wide variety of Conference on the Future of Engineering simulation styles. The steep learning curve, and Education (CSME-ICFEE 2003), on CD, the lack of relevant and readily available exam- 2003. ples, are limiting factors for deploying the HLA in an academic project with tight time constraints. [2] Salustri, F., “Summary of Discussion To help offset this, the first year SYS Team de- Panel on Interdisciplinary Engineer- veloped PoolSim, a real-time simulation of a ball ing programs”, International Con- rolling on a pool table, as a learning exercise. The ference on the Future of Engineer- PoolSim approach to real-time was reused while ing Education (CSME-ICFEE 2003), developing the first year SIDFreD simulator, and http://deed.ryerson.ca/ fil/t/ICFEE02panel/transcript.html. thereby reduced the number of technical issues [3] Serapins, G., “Wish List”, Private Commu- encountered. The second year SYS Team famil- nication, Jan. 20, 2004. iarized themselves with the HLA by extending PoolSim with additional functionality. Again, the [4] Straznicky, P.V., Kind R.J.,“Trends in learning experience greatly simplified their subse- Aerospace Engineering Education - Meeting quent step into the SIDFreD environment. the Industry Needs Using a Team Project”, Canadian Aeronautics and Space Journal, 6 CONCLUSIONSAND FUTURE Vol. 46, No. 2, June 2000. WORK [5] IEEE 1516-2000., Standard for Modeling In this paper we have described the development, and Simulation for High Level Architecture or design if you will, and implementation of the (HLA), IEEE Standard, 2000. 4th year capstone design projects in the Depart- ment of Mechanical and Aerospace Engineering

CSME 2004 Forum 9