NIGHTSIGHT Pan and Tilt: Design Process

"I don't think you can design anything just by absorbing information and then hoping to synthesize it into a solution. What you need to know about the problem only becomes apparent as you're trying to solve it." -- Richard McCormac

Our design process is built on the premise that design is both pragmatic and intellectually informed. Towards that end, we applied several design methodologies which served as a framework to develop abstract conceptualizations and translate them through active implementation into a working prototype. The iterative cycle of prediction, design, and reflection as seen in Figure 10 represents the paradigm which drove our design process.

FIGURE 10. Design Process

Theoretical framework as an integral process.

The prediction phase is the most conceptual of our three part design process and is based on generating questions and gathering information. This phase embodies a mixture of both systematic and creative tools including competitive benchmarking, user interviews, brainstorming, and morphological analyses.

6.1 Competitive Benchmarking

Though previously maligned as industrial espionage, competitive benchmarking provides a means of acquiring a knowledge base of products and mechanisms with functions related to the design problem at hand. The implicit assumption is the precept that effort should not invested in duplicating, but rather, in refining. Benchmarking is a valuable interdisciplinary tool for exploring and analyzing pre-existing systems.

Surveillance cameras, for example, are typically mounted on a pan and tilt mechanism which allows them to scan a scene. Similarly, classroom mounted video cameras such as those used by the Stanford Instructional Television Network (SITN) employ a double axis gimbal for pan and tilt capability. By investigating this related technology, we developed valuable professional contacts like Aubrey Harris, Chief Engineer at SITN, who explained how the current classroom camera systems operate. By looking at such systems, we gained insight into how traditional pan and tilt mechanisms function.

Visiting Hughes' Electro-optical Division in Los Angeles also proved to be a valuable experience in competitive benchmarking. Working in parallel with our efforts, Hughes' professional in-house design team is pursuing a conservative pan and tilt mechanism, which nonetheless, provided important insights into critical issues and obstacles. It became evident that we should pursue cost and simplicity as the avenues of opportunity in our design.

6.2 Interviewing The User

"Our job is to give the client, on time and on cost, not what he wants, but what he never dreamed he wanted; and when he gets it, he recognizes it as something he wanted all the time." -- Dewys Lasdon

The aim of interviewing users is to gather information and identify circumstances that are relevant to the design situation at hand. Listening to the "voice of the customer" is one of the most vital though undoubtedly the most neglected stages of the prediction design phase. Designers are charged with gathering and analyzing customer desires and requirements to ensure that they are addressed in the finished product. The prime motivation of our user interviews was to uncover assumptions and rationales by questioning specifications. In this manner we are challenged to give the customer what they "never dreamed they wanted" --- by recognizing that this may be quite different from what they asked for.

Our visit to the Stanford Sheriff's Department proved to be an extremely enlightening experience. By riding along with several officers during the night shift, we gained insight into standard surveillance techniques and even had a chance to perform a mock surveillance in both urban and rural settings. By interviewing police officers in their working environment, we were able to encourage them to describe and demonstrate aspects of their jobs that could be improved by the NIGHTSIGHT system.

It became clear that the user interface deserved special attention. The officers felt that a knob would give better tactile feedback than a joystick. They expressed a need for an interface which allows for both right and left handed users, and placed a high priority on discreet mounting of the interface given typical surveillance situations. Finally, despite an occasional sentiment that the interface should be simple and very intuitive, most officers quite plainly stated that "cops love gadgets."

Regarding the pan and tilt mechanism itself, it became apparent that although Hughes did not specify a maximum noise level during operation, we felt it would be crucial to design a quiet mechanism for surveillance situations. The officers also explained that since the average life of police vehicles is only 12-18 months, a system which is lockable yet easily detachable would be preferred. Such a flexible design would also facilitate transfer of the NIGHTSIGHT system between vehicles given that police vehicles typically require frequent servicing. Finally, the officers underscored Hughes' desire for a robust and rugged mechanism since police vehicles are exposed to extreme driving conditions as well as occasional vandalism.

In retrospect, our Ride Along experience with the Stanford police was a pivotal event in our prediction design phase. The experience yielded more than just answers to the questions we had brought. In fact, the visit provided insights to potential issues of police procedures which officers may no longer be aware of because they have successfully adapted to these aspects. For example, every U.S. police car is equipped with a movable spotlight mounted on the driver's side A-pillar. Although the current system is extremely counter-intuitive to manipulate, most, if not all, police officers have adapted to the awkward system. In developing a new control interface, it behooves the designer to discern the difference between a good design and a bad one to which the users have adapted. Our goal is to design an interface which considers the current spotlight manipulation device yet is more ergonomic and intuitive.

6.3 Brainstorming

Brainstorming is a means of stimulating a group of people to produce a large number of diverse ideas very quickly. The following list outlines Nigel Cross' five essential rules of brainstorming:

No criticism is allowed during the session.
A large quantity of ideas is wanted.
Seeming crazy ideas are quite welcome.
Keep all ideas short and snappy.
Try to combine and improve on the ideas of others.

Although we implemented the brainstorming methodology during several stages of our design cycle, we found this idea generation technique to be most effective during the earlier stages. In addition, Mind Mapping and List Making, graphical versions of brainstorming, were used and proved to be equally fruitful techniques for ideation.

6.4 How - Why Diagram

The How-Why diagram below demonstrates how we analyzed and dissected the problem of developing a scanning device to be incorporated into a thermal vision system. Why not a searchlight? Considering that Hughes is hiring us to develop a device that incorporates their NIGHTSIGHT camera this is probably not a very viable solution. However, since searchlights must be able to scan a scene, we were led into researching searchlight manipulation mechanisms. This research, in turn, affected our design process and resulted in an additional design alternative --- the gimbal ring design.

FIGURE 11. How--Why Diagram

Shows the free-form brainstorming that occurred when we asked ourselves the questions "How" and "Why."

6.5 Morphological Chart

A Morphological Chart is used to generate a complete range of alternative design solutions for a product and hence to widen the search for potential new solutions. The morphological chart is the result of the axiomatic division of the three critical functions --- pan, tilt, and user interface. We found that the tilt function offered the greatest opportunities for innovation.

FIGURE 12. Morphological Chart

Axiomatic Division of the three primary functions.

From the Morphological Chart, we decided to pursue four designs that were a combination of the following functions:

TABLE 7.  Designs Generated from Morphological Analysis
-----------------------------------------------
Design   Pan            Tilt          Interface  
-----------------------------------------------
Cradle   External Ring  Cradle        Knob       
Generic  Worm           Generic Spur  Joystick   
Bevel    Internal Ring  Bevel         Knob       
Plunger  Internal Ring  Plunger       Knob       
                                                 
-----------------------------------------------

During the prediction phase of the design cycle, many concepts were envisioned. Some designs rapidly proved to be untenable yet others showed promise, and underwent further exploration. The failed designs however were the most important part of our early design process, often leading to "sudden realizations" which served as creative springboards to new designs. Driven by the vision of a mechanism that was user-centered, inexpensive and innovative our design process moved from conceptualzation to rigorous implementation.