Cognitive technology uses in 3D printing and 3D scanning

I am not aware of any applications where cognitive technologies have been used in the new disruptive technologies of 3D printing and 3D scanning. It does not mean that someone is not doing it somewhere and it also means to me that there are huge opportunities in this field for combining these technologies together to create something even more revolutionary and disruptive to change the future of additive manufacturing.

Application 1:​ Improving 3D printing quality of the printed item

I will focus on FDM (Fused deposition modelling) 3D printers. These are the most popularly used printers and I think they will have the most impact on the additive manufacturing world. The other technologies like SLS (Selective laser sintering) and SLA (Stereolithography) are used in the higher industrial and expensive consumer markets and they can also benefit from cognitive technology.

Opportunity:

There are a few problems or issues with FDM like getting consisted print quality. Some factors that influence it is the control technology currently used like stepper motors with no feedback to the controller as well as inconsistencies in the printing material (also known as filament) used for the printing.

There are also a lot of restrictions and limitations inherent in the FDM process, for example items cannot be printed with overhangs of greater than 45 degrees without support. Then there is layer adhesion that is influenced by the material physical properties and the operating temperature (extrusion temp) of the printer as well as the cooling times and environmental temperatures as well as layer height used for printing. These all have an influence on the strength of the final product.

Cognitive technologies can be used in a few different sub areas to improve the overall print quality of the final product. Let’s just focus on 1 area, printer optimisation to get the best possible prints.

Strengths:

There are a few fixed variables and settings that can be used to get the printer setup in the ball park and then with measurement of a few variables like filament width and extruder and bed temperatures other variables like print speed and cooling curves can be obtained for optimal printer setup and max speed to get the best product with the least time taken to produce the product.

The printer can be dynamically controlled with feedback from computer vision looking at the printed items itself with something like thermal cameras to identify optimal layer adhesion to obtain max strength.

 Weaknesses:

This will probably increase the cost of the printer making it less affordable and this was the biggest drawing card to many 3D printer owners.

 Threats:

3D printer owners will reject it because of price

 Conclusion:

This will be a tightrope between improvements in print quality versus cost.

Application 2:​ Using cognitive technology to improve scanned human models

I have been using an Xbox Kinect to 3D scan persons with software called Scanect. There is currently a problem with the lower end affordable software to convert these scans into successful 3d models that is ready to be instantly printed. Almost all current scans need to be altered by software to make these models printable. Things that influence the scans is environmental lighting and speed of scans and colours and textures scanned.

Opportunity:

I think incorporating cognitive technologies into this software with a big database of scanned persons will improve the output to get better 3d models ready to be manipulated and 3d printed.

Strengths:

Scanner hardware resolutions has improved drastically and thus making it easier to get good quality imaging info but the processing software still lags behind to convert into 3D models. Cognitive technologies can be used to fill in gaps in the scans or to find minor errors and fix them automatically with the least human intervention as possible.

 Weaknesses:

This requires a lot of processing and will probably be making the process slower. The use of cognitive technology might also increase the sale price of the software.

 Threats:

The software will become unaffordable to the general public.

 Conclusion:

The ideal would be cheap and affordable scanning and printing of say the typical family holiday moments on the home 3D printer. The trick is to get it done and to make it affordable for everyday use.