Collaborative robots are an area of growth in industrial automation. In some cases, collaborative robots — or cobots — are only collaborative in that they have sensors and safety features that make them less dangerous to human sin close proximity. In other cases, the robots may work together with humans who place items, receive items, or even work together with items. This second type of cobot is increasingly imagined as a humanoid robot with an identifiable head and eyes. But do cobots need eyes?
Why would cobots need eyes?
The question is not about cameras or robotic vision. It’s about whether robots need to have the appearance of human-like eyes. Baxter and Pepper are examples of cobots with eyes. New examples in development sometimes intentionally use eyes to communicate the robot’s intended movements.
New research finds that the eyes aren’t actually that helpful. Robots with eyes seemed more trustworthy to humans who are collaborating with them, but the performance of tasks was actually better without the eyes.
The study in question used a small sample size of 38 people, but checked heart rate and pupil size rather than just asking subjects how they felt. It also considered the amount of time needed to complete the task. Further research might help to determine how much difference the eyes or lack thereof made.
Problems with eyes?
- Equipping cobots with eyes can increase their cost and complexity. This can be a significant consideration, especially for small and medium-sized businesses. It’s essential to weigh the potential benefits against the added costs.
- Where there are eyes there are often cameras as well. The use of cameras and visual sensors in collaborative robots raises privacy and ethical concerns, particularly if these robots are deployed in areas where sensitive information or individuals’ privacy is at risk. Proper safeguards and policies should be in place to address these issues.
- It’s good to consider these matters when designing robots. This may be particularly important as humanoid robots become more popular — at least among researchers and robotics designers.