The objective of the project is to develop a co-robot cane that may be used by a visually impaired person for wayfinding. The device consists of compute vision module, robotic guide module, and human-device interaction modules. The computer vision module includes 6-DOF SLAM for device pose estimation and mapping and 3D object recognition. The robotic guide module steers the cane into the desired direction of travel if the robotic guide mode is selected. The human-device interaction module includes a speech interface that allows communication between the cane and the user and a human intent detection interface that senses the user’s intent about the desired mode-–conventional white cane mode or robotic guide mode-–and automatically selects the mode for the user. This project is co-funded by the National Institute of Biomedical Imaging and Bioengineering and the National Eye Institute of the NIH under award R01EB018117.
Source: National Eye Institute (NEI) // http:www.ualr.edu