Taking it to the Edge: processing for smarter motorcycles

These days, computing is everywhere. From the tiniest devices such as IoT sensors, to phones and watches, TVs and satellites – all the way up to the powerful super-computers that drive scientific research, space travel, weather forecasting, and, more recently, cryptocurrency mining.

No matter their size, all these devices are constantly “doing something”. They’re either acting or reacting and, even when they’re simply idling, they’re still doing something.

At Savic Motorcycles, we’ve taken this concept very seriously and decided to utilise a new computing paradigm called Edge Computing.

Edge Computing is a way of processing data closer to where it’s created or needed, rather than constantly sending it to a centralised location like a data centre or the cloud infrastructure.

Think of it like this: instead of sending the data from your smartphone to a remote server to be analysed, with Edge Computing some of that analysis can be done on your phone itself. This means you can get quicker responses, save bandwidth, and reduce latency for critical decision-making processes.

Edge Computing is particularly useful in situations where real-time data processing is needed, such as in self-driving cars, smart homes, industrial automation – or, in our case, the C-Series motorcycle.

A new paradigm

As you know from our previous blog, all our vehicles are equipped with hardware and sensor components that continuously generate valuable signals and data.

To make use of Edge Computing, we’ve deployed an Edge Computing runtime environment, which allows us to build, maintain, run and operate our ‘edge components’ – the edge-hosted applications and services that perform specific jobs within the C-Series.

For example, we have a component called a Proximity Start Sensor, which determines how far you, as the C-Series owner, are to your vehicle. The sensor’s job is to determine your device’s unique signature, securely connect to your device, calculate a Received Signal-Strength Indicator (measured in centimetres), and determine if the vehicle can be ridden. There is obviously a process of securely binding your phone to your vehicle, but that is outside the scope of this blog.

Another extremely valuable component that we call our Evaluator uses Edge Computing to continuously monitor signals generated by our hardware and software, and determine whether these signals – as standalone values, multiple values in a time sequence, or over a period of time – are within the normal operating range. For example, if one of the cell groups within our battery pack exhibits an abnormal voltage range, the Evaluator’s job is to:

·   determine this by satisfying rule conditions;

·   generate appropriate translation of those conditions, into a structured and well-understood message; and

·   send that message onto the edge runtime message bus (a sub-system that connects various components, which may or may not want to react to those messages).

A different component may then want to do something with that well-structured message. In our case, it may take one of the following courses of action:

·   display a human-readable message on the C-Series instrument cluster display;

·   send a control signal into a receiving component or piece of hardware; and

·   send out an IoT message, which we at Savic Motorcycles can proactively react to.

Edge Computing is very powerful, and opens up a myriad of interesting use-cases that end up as part of current or future customer features – all with the primary principle of improving your safety and rider experience.

In one of our next blog posts, we’ll cover how the cloud (and cloud computing) forms a critical part of our computing infrastructure – and its affinity to Edge Computing.

Watch this space, riding geeks!

Did you enjoy reading this blog?