@sknudstrup gives a clear overview of reference-frame transformations, covering Euler angles, spherical coordinates, and the differences between left and right-handed conventions in 3D modeling.
I’d love to see these concepts applied to force/touch sensors on a robotic hand. It’s an interesting problem - blending a variety of other sensors to provide proprioception and the force/touch sensors to enable object detection and orientation (and feedback to control loops). Vision alone is not going to enable next generation flexible robots. Ultimately the force/touch input sensors are just a 3D/spatial array offset by the robot’s motor positions and joint geometries.
Force sensitive resistors are used for the drumheads of some electronic drums, among other things. They’re printed with a resistive ink on two face to face plastic sheets. More contact equals less resistance. Easier than capacitive because you could potentially print them at home. Two sets of lines at 90deg provides texture data where they cross in the grid.
This stuff is going to be a blast. There’s always more than one way to solve a problem.
Back in the day along with an engineer colleague I co-founded an educational neuroscience startup called NeuroTinker. We developed some hardware (called NeuroBytes) and associated curriculum and neural circuit design ideas for students, makers, and tinkerers of all ages. We used the STM32 platform for our products, but the FSRs can theoretically be integrated into RasPi / ADC circuitry and provide touch like input for Monty.
There are FSRs that are printed, and could be silk screened or printed by hobbyists. They’re used in electronic percussion and can be set up as an addressable grid, making position, area, and pressure detectable in multiple locations simultaneously.