Commodity Virtual Reality (VR) headsets have begun to reach the market, making VR technology more affordable and easier to purchase, set up, and use by consumers. While most consumers purchase VR headsets for gaming purposes, engineering companies are starting to realize their potential use in design, manufacturing, and training. The VR training platform for manufacturing is currently being developed to cut time needed to train employees, increase task comprehension, and minimize nerves that new employees face when training on a real assembly line for the first time. The goal is for new employees to train in the virtual environment before training on the physical manufacturing assembly line.
VR training offers unique benefits to companies that save time and money in the long run. Watching videos and reading paper instructions does not allow for active participation by a trainee in the task they are learning, but the virtual trainer guides the user actively through each step at the user’s own pace. The ability to mock motions from the real world in the virtual environment is likely to also help the user build muscle memory. Onsite assembly line training does have its advantages, but likely requires more time to set up, can be overwhelming to a new employee, requires lots of expensive resources (space, time, money), and may include physical dangers associated with the real task. VR training allows manipulation of 3D models of the assembly line and other required equipment with no need for duplicate parts. It can also warn the user of mistakes without risking physical harm to the employee or equipment.
To evaluate the benefits of commodity VR hardware, a VR trainer was developed using the HTC Vive VR headset. The Vive utilizes two tracked hand controllers and a tracked Head Mounted Display (HMD) to allow users to interact with the fully immersive virtual environment. The virtual assembly line currently features instructional guidance, functional machines with fully intractable interfaces, and visual replication of the real manufacturing assembly line setting. The virtual environment is being developed in Unity.
The assembly line has multiple stations where critical assembly tasks are performed. A trainee performs these tasks by interfacing with more than one piece of equipment. There are hand written quality checks as well as touch screen interfaces and computer screens to track production rates and quality. The worker also must pick up objects, open doors, and push buttons that determine whether or not the assembly line is running or stopped.
The formatting of the training can be separated into four phases. First, users are introduced to VR and basic interaction techniques via a VR tutorial. It serves as an introduction for any users who have never experienced VR or worked with the VR trainer. The next phase allows for the trainee to practice the techniques learned in the VR trainer while learning how to operate the assembly line for which they are being trained. This phase introduces the full virtual environment and allows the user to train on a specific task with full guidance. The guidance system includes step by step instructions displayed by every station in the trainer, beacons indicating intended navigation within the virtual environment, and error notifications if mistakes are made during training. The third phase retains some guidance, but certain layers of instruction will be removed. They will remain hidden unless the user becomes stuck, at which point they will be able to access the instructions on an as-needed basis. The final phase of the trainer will remove all guidance for the user. The entire virtual environment will be open for users to interact with and complete the task from start to finish by applying what they learned in the previous phase. After completing the final phase, new employees will be able to move forward with further training on the manufacturing assembly line with greater confidence about how to perform their tasks.
Throughout each of the phases, data can be collected regarding the number of user errors, error types, user success in specific areas, and overall completion time for any task given in the VR trainer. Any interaction made by the user in the virtual environment can be tracked to help focus on individual improvement. This data can also be used to help improve clarity in the trainer itself.