Development of a user interface for complex terrestrial mov - PowerPoint Presentation

Slide1

Development of a user interface for complex terrestrial movement

Ben WaidaMentored by Craig Schell

Introduction

Any vehicle with three or more degrees of freedom requires an equally complex user interface or control mechanism. This often forces a user to

require

extensive training in order to operate the vehicle safely and effectively. A more intuitive approach to a user interface with three or more degrees of freedom is needed in order to shorten or eliminate the time required to learn how to operate the vehicle safely and effectively. The purpose of this study was to create a more intuitive user interface that enables the user to effectively control a vehicle with three degrees of freedom after minimal exposure to the device.The developed interface must control forward and reverse movement as well as rotation and strafing left and right. In addition, it should allow a user to maneuver comfortably in an environment with little operating space such as a small hallway or corridor. In such tight conditions it is important to note the intended user’s attitude, experience, and motivation when controlling the vehicle (Nuance, 2009).

Materials and Methods

The CAD program SolidWorks 2014® was used to create dimensional models for each component of the interface and compile them into a full assembly (Figure 1). Each component was machined in aluminum and assembled using steel dowels and set screws. When the body was completed, three hollow shaft Vishay® encoders where attached to the dowels in order to measure the rotation of each component of the interface (Figure 2). Each of the three encoders were wired to an Arduino Uno® which read the voltage input and sent the information to a system running LabVIEW™. This program would then control the motors on the test vehicle, allowing the rotation of the interface to correlate with movement patterns in the vehicle which only had two degrees of movement, forward/backward and rotation (Figure 3). Through slight variations in the program and the location of the interface, four different versions of the same interface were created with different axes affecting different movement patterns.In order to test each variation, 32 Science and Mathematics Academy seniors were split into four groups of eight, one for each interface variation. A course consisting of a right turn leading into a left turn was laid out using tape. Each student completed the course both forward and backwards using their group’s respective variation, having the time to complete each segment (Graph 1) as well as the number of errors made recorded. Upon completion of the test, each subject filled out a questionnaire where the interface was rated on a scale from one to five on ease of use, forward control, backward control, learnability, and sensitivity (Graph 2).

Figure

1 (top left): CAD assembly of the full user interface.

Conclusions

The effectiveness of each user interface can be determined based on the amount of time each person took to complete the course as well as the reviews of each test subject. The aspect importance data can be used to determine the appropriate weighting for each category. Based on these factors it was determined that, in the given scenario, the interface placed over the wheels and with the yolk(roll) for steering was the most effective as it had the highest ratings for both ease of use and forward control, the two aspects deemed most important.

It was originally believed that the interface located on the casters with the grip steering would be ideal due to its overall decent ratings, especially with its high backwards control. Upon completing the aspect importance survey, this interface lost much of its value as backwards control was rated the least important of the attributes. The reason for this rating was likely due to the lack of complex obstacles on the course, which would make it more difficult to traverse forward versus backwards as the vehicle would block vision of them in the forwards direction. It was also determined that the ring component of the interface should be made larger or eliminated as a user’s hand could become stuck if control of the vehicle was lost while moving backwards.Further studies could include user interfaces that utilize types of motion other than rotation such as applied force or translational movements. In addition to this, a more complex course could be created in order to better evaluate the importance of various aspects and how each interface is able to handle more complex maneuvers. More intuitive user interfaces will shorten the training time on complex vehicles allowing users to operate them more efficiently.

References

Nuance (2009). Attitude, Experience, Motivation and Key Issues.

Automotive Voice UI Usability Study User Survey, 1 – 16. http://www.nuance.com/ucmprod/groups/corporate/@web/documents/webasset/nd_002994.pdf

Graph 1: This graph displays the relationship between each variation and the average amount of time needed to complete the course forward and backward.

Graph 2: This graph displays the average score each variation received amongst the five categories the interface was compared on.

Graph 3: This graph displays the average score allocated to each of the five aspects in order to determine the importance of each aspect in a successful user interface.

Figure 3 (lower left): Machined and assembled user interface with encoders attached and no handle or mounting box.

Figure 2 (right): Vehicle used for the study of the user interface with two possible interface locations.

Results (Continued)

Each interface was found to be similar in terms of forward control with the on wheels and yolk steering variation being rated slightly higher than the others (Graph 2). The on casters with grip steering variation was rated to have the best backwards control (Graph 2) which is further supported by the average backwards completion time of the variation being significantly shorter than the other variations (Graph 1). When looking at aspect importance, ease of use was ranked as the most important aspect with backwards control as least important (Graph 3).

Results