On the Road…with UniPhi

I must say, the Tesla was a dreamboat. Roughly 300 miles per charge meant that I didn’t have to force myself to do a 500-mile stretch based on a tank of gas. I would make it to a charging station, stay within distance overnight, or do a 30-minute charge and be off. It was fun to think that I drove 9,718 miles with a $0.00 gasoline bill!

I used the self-driving mode pretty much anywhere if the roads were marked.  And driving through the plains, where a lot of the roads are long stretches with very little traffic, while still keeping my eye on what was coming my way, autopilot was really quite relaxing.

The Tesla also has a cool feature where you can actually fold down the seats and create your own, quite comfortable, bed. On those nights where I wanted to gaze at the stars, or my charging station was way out in the wilderness, I was able to enjoy my Tesla-BNB.

Thanks to connectivity through the vehicle and my phone, I was able to continue making progress on work assignments along the way. Here are some of the things that truly caught my attention:

Spreadsheet errors that were thought to be correct, but weren’t.

Embedded software development can be so complex, often combining millions of pieces of data that have to align without error. It is essential that, before something goes into production, it has been proven that the software is error-free. However, even with extensive manual checking, errors can still show up.

One of our clients related this story. An intern in one of the groups that was not using our tools gave a presentation regarding spreadsheets with 1,000s of variables that are intended for production applications. Having heard about UniPhi, he used the tool, and consequently was able to find approximately 500 errors in those very spreadsheets. With a few clicks, he was even able to take his demonstration one step further and automatically mitigate these issues. He then showcased this to the management team. “Let me show you. I’ve solved this problem of inconsistency by using this tool.” (He was able to do this just by simply importing spreadsheets into UniPhi and using its powerful automated checking and mitigating capabilities.)

At least two managers who had come into the meeting unconvinced that UniPhi could be of support, came out of the meeting saying they needed to begin implementing its use.

It’s like the analogy of when people used to use a horse to get around. At that time, what would they say about improvements when confronted with the question, “How could you improve this method of transportation?” Well, they could say that they needed a faster horse or a wagon that could pull more. They never would have thought of a car with 200 horsepower—much less an electric autonomous vehicle that can accelerate to 60 mph in 4.1 seconds. It’s often the case that we become so ingrained in the status quo that we don’t stop to think about the problem from a different point of view.

Likewise, people who are accustomed to using spreadsheets will say that they want a better spreadsheet. They haven’t considered intelligent automation that automatically checks and resolves errors and inconsistencies, or points them out for consideration.

Application of Custom Rules in UniPhi.

At night during my trip, I’ve been diving into the functional safety standards for Automotive ISO 26262. It’s fair to say that the game has changed because of this standard. This is a set of rules that you can’t ignore. It applies to all automotive systems; but recently, it tends to key in on electrification and ADAS – advanced driver-assistance systems. And inside this world, you have a lot more complexity.

But safety standards are not just for automotive. While safety requirements are modified by industry, all industries have similar safety standards.

The challenge is how to satisfy them? Which is the best path to satisfy the hundreds of rules required by these standards? You can try to do this manually. And many do right now. But this is very time-consuming and error prone.

With UniPhi, the tool would be checking everything in real time to make sure that all the developers are able to follow the rules consistently. This is all done in an automated fashion. It’s also a robust method for teams to incorporate their own custom rules. Through UniPhi, everyone would be alerted to the presence of built-in and custom rules.

This is somewhat akin to my birding. Sure, I could find the habitats manually in a book of birds by state. But I was so much more effective with the app I have that automatically indicates where particular birds had been sited the day before, or even that same day. Rather than spending my time searching vast areas of suitable habitat, and hedging my bets, I could focus my precious, limited time on actually observing the birds. This was possible only because I eliminated the need to take all of those tedious, time-consuming tasks that would be required without the use of technology that simplifies the process and leads to results.

And if I broke the rules and got too close to a moose, my Tesla would get me out of there in 2.4 seconds…while also alerting me that going 60 in Yellowstone would break other rules that are in place to keep me out of trouble.

Legacy Model Imports Recapture Valuable Assets.

Clients who immediately benefit from our tools are innovators from around the world who are already using the Mathworks tools for model-based development. But to benefit from UniPhi, you don’t have to be using these tools. In fact, you could just be hand coding. Or you may have legacy models that you would like to make use of.

Importing models into UniPhi is no problem. Where the challenge lies is that in starting to work with these legacy models, the user usually has to resolve many inconsistency issues that break numerous rules of data and architecture.

But with UniPhi, in addition to built-in rules you can define new rules; then the tool automatically detects errors in the legacy models. With UniPhi’s intelligence, the machine either automates or guides the user to error correction, so you now have an updated, error-free “legacy” model.

In this way, you can incorporate valuable assets from the past that still have relevance to the work you’re doing now. Again, it’s a bit like my Tesla. I can stay in standard driving mode and recapture the feeling of hands-on-the-wheel driving passion, or move to a new model, hands-free autonomous, that allows me to focus on other things while the machine senses what’s around me and either compensates automatically or alerts me that I need to come back on line.

And the beauty of it is that all of us can rest easy. All of our clients with UniPhi.

And me…reclining…looking up at a starry, starry night.

Get Connected

They say, “Birds of a feather, flock together.” If you are in embedded software development and you’ve felt a need for something more when it comes to developing error-free embedded software, connect with us for a free consultation.

Or, if you enjoy birds and would like to see some of my trip pictures…

Send me an email. Let’s talk.

John Mills