6 Replies Latest reply on Dec 3, 2014 9:30 AM by userc_6396

# WICED Sense

I am looking to use a WICED Sense for movement, does anyone here know if a movement accuracy of 10cm or less is possible?

Thanks

• ###### 1. Re: WICED Sense

i think the onus is on the sensor or transducer that is supposed to detect or measure a "movement accuracy of 10cm or less". then we can make use of bluetooth to transmit this data to say a smartphone. what is this device that you have in mind?

• ###### 2. Re: WICED Sense

Hello emonster,

Here is the WICED sense Schematic: WICED Sense Reference Design Schematic (PDF)

You will find the ST Micro Part numbers of the sensors and can find the datasheets here: MEMS and Sensors - STMicroelectronics

Thanks

JT

1 of 1 people found this helpful
• ###### 3. Re: WICED Sense

Hello boont,

This is for a research project, the device is for exercise tracking, for example weightlifting...

- Bench Press, we need to track the distance, acceleration, velocity and time of each rep.

So we ideally need an accuracy of 5cm or less... 1~2cm !!

Thanks

Enrique

• ###### 4. Re: WICED Sense

WICED Sense has an accelerometer, gyro and a magnetometer. It does not directly measure displacement, but it may be possible to get a fair idea of the displacement with respect to the position at the previous sample point using the data from these sensors. This is really something the app will need to compute,

• ###### 5. Re: WICED Sense

It seems that others have approached the problem of measuring displacement using an accelerometer: http://cmis.mis.brighton.ac.uk/staff/gw4/papers/Displacement%20from%20Accelerometer.pdf

• ###### 6. Re: WICED Sense

Using accelerometer and gyro to "predict" direction and the "speed" of movement might be possible.

But 10 cm?

I would guess the resolution of the sensor is not good enough: imagine: to move an object just by 10 cm will result in very low acceleration and very small interval where sensors have to realize and track the forces/acceleration.

So, I think, the success depends on: a) sensors sensitive enough? (and you had to calibrate anyway) b) checking sensors can be done fast enough (tracking 10 cm distance for start, stop and samples in between to get acceleration might result in MHz sampling rate) c) if the MCU fast enough to do? (24 MHz core clock, I2C/SPI slow speed interfaces to sensors, several register accesses through it ...)

Tracking a person in larger room, maybe with accuracy of +/- 5 feet (1 m) over a time of several seconds might work. For objects moving 10 cm (in ms time interval) - I am not convinced.

Best: check such research studies, master thesis for accuracy of their solutions/implementation.