So one of the main objectives of my PhD research was to achieve the difficult task of developing a learning algorithm for machine learning ( RBF neural networks to be more specific) applications, that would enable the prediction of drone propeller damage in real-time AND without altering the bought-out flight controller ( DJI Naza , APM , Pixhawk , etc...). The only way it could achieve that was by analyzing the outputs of the flight controller sensors and learn when a fault would occur. Well, I believe I'm getting closer to this objective (submission is Nov 2019). I've decided to include the two figures below which illustrates the training process (0.2 sec on desktop) and prediction time (0.008 sec) and the accuracy to the true dynamics of the quadcopter drone. In this case the pitch dynamics are being predicted. Although noise hasn't been introduced, it's quite clear from the graphs, that the learning algorithm has enabled the RBF network to accurately capture th...

Don't get me wrong. I like Mathworks products, specifically MATLAB and the suite of control toolboxes. But for curse of being a super monopoly software is that, in order to make profit to drive software updates (which is now expected to be at least twice a year on most toolboxes), the focus is on the popular user requirements. This is often based on academia and research houses which often doesn't deal with hardware or firmware. In so doing, a large percentage of professional engineers productivity is directly coupled to the level of expertise of the analysis tool rather than the process upon which the product is based. Consequently, the poorer or bloated the tool, the slower can an engineer develop a product at lower cost. Unfortunately, the reliance on more powerful machines is the way to circumvent to problem. Just run the bloated simulation fast enough that's it's on longer noticeable to the designer. But is that engineering design? Is not the whole point to i...