RetroFlex : enabling intuitive human–robot collaboration with flexible retroreflective tags

Efficient and seamless human–robot collaboration relies on carefully-designed user interfaces to relieve the users from heavy mental stress and burden. Current marker-based intuitive interfaces suffer from reduced reliability in a noisy environment, non-scalable wireless connections, and fixed form factors with limited data capacity, which limit the efficiency of real-world applications where frequent and seamless target switching is required. We propose RetroFlex , a flexible retroreflective communication system that enables fluent and intuitive human–robot collaboration. RetroFlex integrates robots with a flexible tag (FlexTag) and leverages visible light backscatter communication to exploit the intrinsic user spatial context to retain the intuitiveness in the interaction process. The users can interact with multiple target robots in a point-and-control manner, which significantly lowers the workload and improves the overall experience. Our evaluation reveals that RetroFlex is able to support room-scale tasks with commodity smartphones with a bit rate of 60 bps at a distance up to 2.5 m and a view angle up to 70 ^∘ while being robust to different environmental noises. A usability study with 12 users and two real-life tasks demonstrates that our system offers a fluent and satisfying multi-target control experience.