The Guide Dog Robot: Future Mobility Aid for the Blind and Visually Impaired
Independent travel is one of the strongest desires of about 285 million people who are visually impaired in the world. Since the concept of technological assistance of the blind is of recent origin, they have been largely depending upon more sensitive and subtle utilization of their remaining senses. Many ways and means have been used by visually impaired persons to satisfy the basic desire to be mobile. From earliest times, animals, sighted people, and devices (such as sticks or canes), have been used to achieve varying levels of mobility. Basically, there are three (possibly four, if electronic travel aids are considered) common ways of getting about. The first is with the use of a sighted human guide. The other two most accepted and proven methods are with canes of varying lengths and dog guides.
Ideal mobility aids for the blind should support the three necessary functions for mobility; i.e., (1) the blind persons next step, (2) his/her directional orientation, and (3) his/her navigation along reasonably long travel path on both familiar and unfamiliar paths. However, there are lots of existing devices such as the path sounder, the sonic glasses, and the laser cane which only have the first two functions. The information processing system employed by the existing devices is very simple and crude so that the blind user must concentrate on the devices, resulting in the fatigue of the user or loss of their information which otherwise might be obtained through the remaining senses.
As a new technology developed for enhancing the support for the visually impaired person, guide dog robot has reached to a new level in human innovation aspect. The purpose of the guide dog robot project which started in 1977 is to enhance the mobility aids for the blind by providing them with the functions of guide dogs .i.e. Obedience in navigating a blind master, intelligent disobedience in detecting and avoiding obstacles and well organized man machine communication which does not interface with blind persons other senses.
In its simplest form, robot should be able to avoid the obstacles and secure its blind owner from all sorts of harm in their path. But this project by no means aimed at reproducing the guide dog itself completely, but to extract fundamental functions of a guide dog and to realize those functions by a method appropriate to machines. But since the ultimate goal is a guide dog, they called this project The Guide Dog Robot.
Fig 1: Guide Dog Robot
Japanese technology manufacturer NSK have come up with a new concept aimed at helping the blind, a robotic guide dog. On display at this year International Robot Exhibition in Tokyo, the four legged robot with wheels is able to climb stairs and maneuver around obstacles independently. With the ever increasing aging demographic, the robot has been developed to answer the problem of a lack of resources to cope with an increasing demand on the health system. Guide dogs are assistance dogs trained to lead blind and visually impaired people around obstacles. Dogs are very capable guides for the blind and it has been serving the blind not only as guide but also as a good companion through all the difficulties in his/her travel. But they require extensive training. Fully trained guide dogs cost between QAR 20,000 – QAR 50,000 per year for training, breed and support; they are only useful for about five years. Furthermore, many blind and visually impaired people are elders and find it difficult to care appropriately another living being.
The Guide Dog Robot uses a Microsoft Kinect sensor to visually map obstacles like stairs in 3D, while sensors added to the quadruped's feet have helped minimize crashes. It's easy to critique the robot for its lethargic performance compared to a trained Seeing Eye dog, but the potential advantages of such a creation are very exciting. Users hold a pressure-sensitive, variable-length handle on the robot while being guided around. The robot has obstacle and distance sensors in its head and legs to detect stairs. The robot works through force sensors in the handle grip which instruct it to move depending on how it s held, forwards to go forward or pressure on the sides to turn. The robot also processes images through sensors located on its legs which converts information into information in a 3D space.
While voice recognition has been implemented, in addition to simple commands like start and stop, users could one day tell the robot exactly where they wanted to go and a combination of GPS and wireless communications would let it automatically map out an ideal route.
The quadruped beast can also scamper up and down steps with more grace than its predecessors, which moved more like arthropods than actual dogs. The bot's paws have also been equipped with obstacle-avoiding bumper sensors, and researchers are working on incorporating voice commands, as well.
If the Development and production of this robot goes ahead, it can solve a serious need and has the potential to improve the quality of life of many vulnerable people, while saving money. The researchers at NSK have a lot of work to do before their creation could reliably lead a visually impaired user around town, but they're optimistic that the technology could be commercialized by 2020.
- All on Robots n.d, Robot motors, viewed 15 October 2012, http://www.allonrobots.com/robot-motors.html
- Leicester W. Farmer 1978, ‘Mobility Devices’, American Foundation for Blind Inc., pp. 48-111, IEEE Xplore Digital Library, viewed 29 September 2012
- S.Tachi, K.Tanie, K.Komoriya & M.Abe 1985, ‘MELDOG: The Guide Dog Robot’, IEEE Transactions on Biomedical Engineering, vol. 32, no. 7, pp. 256-270, IEEE Xplore Digital Library, viewed 8 October 2012
- The Blind Based on Mobile Robotics Technology,” IEEE Trans. Biomedical Engineering, vol. 45, no. 11, pp. 1376-1386, viewed 13 October 2012
- Iwan Ulrich & Bornstein J 2001, “The Guide Cane- Applying Mobile Robot Technologies to Assist the Visually Impaired,” IEEE Transactions on Systems, Man, and Cybernetics, viewed 10 October 2012, http://www.personal.umich.edu/~johannb/Papers/paper72.pdf
B.E Robotics and Mechatronics ’09
Swinburne University of Technology, Malaysia