GPS is fairly straightforward, however, there are situations where GPS is not available, or the environment is not suitable (heavy forest or tunnels). Triangulation is useful either to improve the GPS navigation or to be used alone when both the positions and the throughput of the radio beacons are well known.
Inertial navigation, also called “dead reckoning”, does away with the need of external beacons of any kind, it is self-contained and can’t be jammed. It uses gyroscopes and accelerometers onboard the mobile robotic vehicle. The accuracy of inertial navigation is determined by the efficiency of the hardware, more precise gyroscopes and accelerometers produce better results.
Indoor navigation presents a somehow similar problem. Of course, standard GPS can’t be used, so an indoor navigation system will comprise of triangulation, inertial and some sort of shape recognition. The easiest way is to use a magnetic strip or colored strip on the floor. That is not actual navigation but just following a predetermined path. An autonomous mobile robot will use visual elements, such as QR labels – that will be the simplest way – to actually identify the surroundings. Triangulation can be used to a certain degree if, for example, the environment is static, rather than an always changing warehouse.
Business Intelligence is something that everybody seems to know how to do, but rarely people understand what it is about. Because the results, most of the time, are sub-optimal, the term has the characteristics of a buzzword. Business Intelligence is not about fancy reports with charts. Business Intelligence is, more than anything, Pattern Recognition. It is the capability to mine, analyze, interpret the data, recognize the patterns and determine the trends. That is what Business Intelligence has to do.
The Importance of Learning
The capability of a system to learn from previous experiences is the main trait of its intelligence. The evolutionary processes in the biological world made the self-learning process the corner stone of survival. Without self-learning, any system is doomed to fail, in one form or another, sooner or later. A self-learning enabled system needs to be capable of recognizing patterns and the effects of these. Such a system needs to be able to make decisions and execute tasks without any prior programming.
Pattern Recognition is the basis of any intelligent system. Without pattern recognition life wouldn’t exist, to begin with. It is the capability of any life form or system to distinguish an edible fruit from a poisonous one, or a dangerous predatory animal from a tame herbivore. It is the bedrock of any system with any form of intelligence. Pattern recognition is related to shapes, colours, but also trends in datasets, from financial data to error logs.
Image processing is highly related to pattern recognition. Whether we are talking about face recognition or car plate recognition, the same type of algorithms are employed, more or less. In the case of facial recognition, a further step is emotion recognition, which is a subset of the same. Other types of image processing are related to identifying the type of objects based on certain characteristics.
Bayesian Networks are a boolean logical statistical application, capable of determining the outcome of certain actions with a certain degree of accuracy. It is similar to the logical analysis that human and any biological life forms do in different ways. One way or another, any life form does some form of risk assessment based on different inputs, previous experiences, and pattern recognition. In AI systems this process is simulated by using Bayesian networks to determine the best course of action. The applicability is vast, it can be used for more or less anything that requires a logical decision-making workflow.
We can design and quickly develop prototypes for mechanical and electromechanical systems together with one of our partner companies. Having access to modern manufacturing capabilities, and along with our software skills, we are capable of developing complete automated and robotic systems for industrial use.
Why do you need it?
If you are a business owner then you need to ask yourself the following three simple questions.
1. Are any tasks in your business currently done by human workers, that are repetitive and boring?
2. Are any tasks a bottleneck to your businesses' productivity?
3. Do these tasks involve physical or visual activity?
If you are remotely interested in automating your business then reach out to us to find out more about the kind of services we can provide.