EMBEDDED TECHNOLOGY


The new concept of continuously adaptive AI


How individual edge devices can improve performance by instituting a feedback loop to learn from their environment


By Tyler Baker VP, Engineering (Qualcomm Innovation Center) – a Foundries.io team member


Lesson number one that AI developers learn is that any AI system is only as good as the dataset it is trained on.


That’s why a lot of the effort ploughed into the development of edge AI systems is devoted to curating the training dataset, so  effective learning for the intended use case,   Today, this training dataset is normally general: a model for recognising cats visually is trained on tagged pictures of   would not go to the locations where its cat-  take millions of images of the local cats in  This approach to AI model training is  the device to understand its operating  in many applications of AI at the edge, this type of universal training dataset fails to provide the device with the knowledge it  Take the example of an autonomous    and weeds and to treat different weeds  weed recognition model, the developer will train it on images of the weeds most   This limited model is then programmed into  


This might work well until a previously unknown weed starts to appear in the  


 take account of local conditions


  the crop plants on which the model was  a weed to thrive while killing the fresh shoots 


 of medicine, personal health monitoring   cash-poor and time-rich: for example,     of physiological data that the device is  


So how to solve the problem of the general training dataset?


The answer might lie in instituting a  training dataset, the model development system in the cloud and the data captured


    io and Edge Impulse is the continuous    smart farming example, physiological data            


 individual environment, the developer would  


JULY/AUGUST 2025 | ELECTRONICS FOR ENGINEERS 19


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46