According to the World Health Organisation (WHO), 285 million people are visually impaired worldwide: 39 million are blind and 246 million have low vision. Additionally, about 90% of the world’s visually impaired live in low-income settings. 

Blind people have problems in walking or navigating independently. As such, aides have looked to solve this issue. The most widely used is a white cane; it is a useful tool, but it has certain drawbacks such as small radius and struggle to detect objects on the ground. The other solution is guide dogs, but it is quite expensive and not easily available in developing countries where most blind people live. 

The main objective of our project is to develop a wearable, cost-effective, efficient and independent smart guidance system that will assist the blind in avoiding obstacles, identifying places or objects and navigating from one place to another. 

We define this as alternative perception. That is to say, using devices to sense the environment and present the blind with meaningful information about their surroundings, allowing them to navigate the area with ease. 

We have decided to utilise the Kinect unit for this project. It is a motion sensing input device developed by Microsoft for the Xbox video game. It consists of an IR depth sensor and RGB camera. The IR depth sensor is used to obtain the depth image data of the environment. This sensor is quite powerful and can provide valuable image data to be processed to extract useful information for the blind people. 

One setback to the Kinect unit is that the floor itself may be detected as an obstacle. To overcome this limitation we have developed an algorithm that can differentiate between the floor and the obstacles. In this way, the floor will not be detected as an obstacle and obstacles of various sizes above the floor can be readily detected. Furthermore, we have added important additional features, such as the provision of 2D printed codes, which can make the indoor navigation for blind people more effective. 

Our solution uses a combination of five vibrators to alert the blind about the obstacles. We occasionally use audio instructions for warning the blind about immediate danger and for providing useful information other than routine guidance.

We are building a low-cost robotic arm that empowers people who currently require long-term care to continue living independently.

In the US alone, more than 12 million people require long-term care due to illness, disability, or advanced age. As they become unable to look after themselves, assistance can include anything from help navigating the environment to completing basic tasks like maintaining personal hygiene and eating. This takes a personal toll on the individuals who suffer reduced quality of life and also has significant economic impact. Currently, long-term care accounts for 9.3% of all personal health care spending in the US.

Since most people requiring long-term care are elderly, our aging population has a huge effect on these numbers. In the US, approximately 63% are all people requiring long-term care are over 65, and 30% have substantial long-term care needs. This trend is rapidly accelerating. By 2050, 1.5B people (or 16% of the world’s population) will be over 65. Caring for this many people and allowing them to retain independence requires significant leaps in our robotic capabilities. With our current technology, we are ill-equipped to face these societal and economic challenges, let alone provide the quality of life that each individual deserves.

Robotic arms are the perfect tools for enabling people to regain their independence. They have long been prized in manufacturing for their unrivalled precision, repeatability, and versatility where they play an instrumental role in making the products we buy, the clothes we wear, and the cars we drive. In fact, anything a human arm can do, a robotic arm can probably do better.

The problem is that they are prohibitively expensive and usually quite difficult to use. The cheapest robotic arms today are still more than $20k and generally require specialized knowledge to perform anything beyond the most basic tasks.

We are also just beginning to see industrial robotic arms adapted to assistive situations. Some companies have recently demonstrated attaching a robotic arm to a power wheelchair to assist people with upper body disabilities. The arm is controlled using a joystick and is capable of opening doors, retrieving items from shelves, and even enabling people to eat and drink without assistance. However, those arms cost 10x more than the chairs to which they attach.

We believe that everyone should have access to advanced automation no matter what their circumstances. This is why we’ve made KATIA. Our low-cost robotic arm that makes fully independent living a reality. KATIA has the capabilities of an industrial arm, but the price and usability of a laptop.

KATIA can be mounted to an electric wheelchair to serve as a virtual prosthetic arm that accompanies the person at all times, helping interact in both home and work environments. KATIA can pick items off the floor, assist in personal grooming, and perform any number of tasks that would normally require assistance from a caregiver. For tasks that are repetitive or out of reach, additional KATIAs can be bolted to stations and operated remotely, so that individuals have even greater control over their environments. 

We’ve designed KATIA to be simple and intuitive to control, regardless of your physical capabilities. At the most basic level, KATIA can be manually guided through a motion that it plays back flawlessly. For direct control in realtime, customers can draw paths for the arm to follow using our mobile application, simulate gestures through peripheral devices like a Kinect and Leap Motion, or operate direct positional control with a 3d mouse. People without the use of their limbs can also coordinate with myoelectric sensors and integrate with voice recognition software to have KATIA perform pre-programmed tasks. For more autonomous control, applications can be loaded onto KATIA’s powerful embedded computer and extended with add-on sensor modules and external integrations.

At Carbon Robotics, safety is our guiding principle and KATIA is the safest robotic arm ever built. We’ve invented a technology that turns the entire housing into a giant capacitive sensor that can detect a person from 0 to 60cm away in any direction, from anywhere on the robot. Our sensor has no blindspots or pinch points and works automatically with no setup or calibration necessary. This means we can reliably detect collisions long before they happen and avoid them altogether.