Intelligent Guide Robots for People who are Blind or have Low Vision: A Review

According to the World Health Organization, at least 2.2 billion people worldwide are blind or have some degree of near or distance vision impairment. Guide Dogs and other existing mobility aids are valuable tools to support travel and independence for these individuals. Nevertheless, the process of training dogs to be effective mobility aids is expensive, time-consuming, and requires highly specialised expertise. With technology making significant inroads into modern society, intelligent robots that are powered by smart sensing and advanced artificial intelligence present an opportunity to improve the mobility of those with blindness or low vision where access to a Guide Dog is not available. However, developing robust and effective robotic mobility aids with static and dynamic settings for use in both indoor and outdoor environments has remained a difficult task to accomplish. This article compares the capabilities of both existing robotic mobility aids, and existing cutting-edge robotic technologies, against a set of fundamental functional mobility aid characteristics with the aim of proposing a minimum viable product to support people who are blind or have low vision in everyday mobility and navigation tasks. The functional qualities identified and examined were sensing and interpretation, obstacle avoidance and object targeting, fluid and adaptable movement and navigation, interlink of device interface and sensing-interpretation unit, device interface to user, form factor and design. The outcomes of our study indicate that while existing robotic aids are limited in scope and focused on a narrow range of functions, existing cutting-edge robotic technologies have the potential to combine to create a fully-functioning robot guide. We propose a minimum viable product using these technologies as the next step towards a fully-functioning robot guide supporting the mobility and navigation of people who are blind or have low vision.