There is a saying “he/she has eyes in the back of his/her head”. This means as much as the ability to see what is happening behind you. With the advent of inexpensive cameras mounted in your car this is a reality. Look at dashcams from for instance Nextbase. Just put one in your car and you can record what is happening in front of you as well as behind you. When autonomous driving matures, we will see more and more cameras being mounted in a car together with all kinds of other technology.
Line of sight
Regular cameras have some drawbacks. They record and do not interpret what is happening. But this means that a camera will happily record the accident that is about to happen, but does not engage in any action to perhaps circumvent it or minimalize the damage. That is not possible since a dashcam is now an aftermarket (installed after the car was built) device and not connected to the car itself (other than power supply) and it of course requires AI / Deep Learning to be able to assess the situation. But this technology is under development and we will see these features emerge over the next couple of years.
Another thing is that a regular camera cannot see around a corner. There is this thing called ‘line of sight’. If there is a car or person around the corner that you should give right of way, it is not possible to see that. Of course, MIT Media Lab has a solution for that, CORNAR can look around corners using femto photography (a femtosecond is one quadrillionth (10^15) of a second). A laser sends out a signal and when that signal hits a wall it is reflected in all directions and therefore when they hit another object bounce back again and get captured by the camera. Very complex, but there is an easier way. Nissan showed their vision at CES with see the invisible
We are going to see more and more connected cars. Not only the car to the internet but also the car to other cars, traffic lights, road signs and so on. With your location, speed and direction you indicate where you are driving. A car does the same just around the corner (out of your sight) and by looking at the data for that specific location we can actually ‘see’ that there is a car that we should be aware of. With autonomous driving, the cars will do this automatically given a set of rules like right of way but perhaps even other rules that take into account other circumstances beyond right of way. You could think of emergencies or rules that will lessen congestion in a certain area.
You think that we are now reliant on ICT technology, and we are. We have seen situations where the unavailability of the 3G network in The Hague, the third-largest city in the Netherlands, stopped the trams from running a couple of years ago. With 5G as an ingredient technology in autonomous driving, we need high availability and resilience to the max. Just to make sure that everything is communicating. Otherwise we have a single point of failure that is easy to manipulate for hackers, foreign governments and other people keen on disrupting our society. This is as much part of the vital infrastructure as power and water is.
But there is also the question of security. When we (or our devices) rely on data sent from sensors and so on, we need to be sure that the data is correct, timely and unchanged. This means encryption, security by design and lots of testing. And don’t forget privacy! Who we are, where we are and what we do is very much personal data and should be treated as such.
Driven by APIs and streams
Since we are talking about the future we do not know exactly how these systems are going to communicate. It could be via APIs or streams of encrypted data that will be interpreted either on a server or even on the edge of the network (in the car or device itself). What ever happens, security is paramount. On the other hand, we must be able to communicate with others in order to move from A to B. So, there is somewhat of a catch-22, to make it safe we should not allow communications but we need communications in order to use it. API and stream Management solutions will grow over the next decade or so to enable more and more of these systems.
Managed and Monitored
The issues I describe in this blog pertain to the future but are true for the current day and age. No company or organization should ignore security. When we look at solutions for managing our current IT landscapes: API Managers, Enterprise Service Busses, Identity and Access Management solutions and Stream processing solutions, we need to manage and monitor them.
The level of management might vary from deployment to deployment but the three principles described earlier apply:
- Only those who should have access should be able to
- The data should be secure, timely and untampered with
- We need security by design.
We propose a state of the art solution in the three aforementioned areas of API, ESB and Stream Processing. With this prerequisite in place for a secure solution you can deploy and build the system that you need in order to run your current business securely.
To learn more, take a look at for example our white paper on ESB. If you have any questions, reach out to us and we will be glad to discuss them.