Zero Clearing Operation of Six Axis Force Sensors

1
When providing technical services to users, we are often asked questions related to sensor reset, which is sometimes referred to as skinning.
The zero clearing of six axis force sensor refers to the user specifying a state as the reference state of the sensor through commands or certain operations, and the sensor’s subsequent data is output based on this reference state.
For a six dimensional force sensor, this reference state is a set of data containing six component force information. For example, when a sensor is placed flat on a desktop, its output is a set of data that is close to zero in all directions. After installation, due to factors such as the end tool, its own gravity, and the angle of the installation surface, its data output is different from that on the desktop, and the force values in one or more directions may be much larger than zero.
2
At this point, users always hope to use the current sensor status as a reference value based on their operating conditions. The data after the sensor is subtracted from this set of reference values. This operation of recording the reference values and recalculating the output results is called resetting or peeling.
In fact, it is easy to understand that users have such needs, because for an ideal sensor, we always hope that it is 0 before the action I am concerned about is generated, and the data increment after the action is generated is the force measurement value we want. But in reality, the sensor is not idealized. Its output is based on the zero position state at the factory, and it is not clear what state it is currently in, whether it is flat or side, so its output is not an ideal set of zero values in the user’s perception of the zero position state. So the user needs to specify a new zero position state, so that all sensor data is output based on this new state.
The reset operation of sensors is divided into two types: hardware and software:
Hardware reset is a command from the upper computer that requires the sensor firmware to remember the current measurement value as a reference value, and subsequent measurement outputs are based on this new reference value.
3
Software reset is when the upper computer software or program records the current value as a reference value, and the data received thereafter is recalculated based on this reference value. The processing logic of these two methods is the same, but the actual effect is slightly different.
4
This is because when we actually perform a zero reset operation, we cannot only take the current value as a reference. The actual force sensor is not idealized, and its data always fluctuates slightly. If we only take the current value as a reference, it is very likely that this reference value deviates from the mean of the past period of time. Therefore, we need to continuously collect data for a short period of time, which is more suitable between tens to two hundred milliseconds, and then calculate the arithmetic mean of the data during this period as the reference value. Just now I mentioned that the difference in the effect of hardware and software zeroing is due to the different time taken to calculate the average value of this continuous data. The embedded processor inside the sensor has a clock speed of only a few tens of megabytes, and its computing power is too poor. It takes one or two seconds to calculate the average value of the data. The upper computer software runs on a CPU with a main frequency of several gigahertz, and it only takes a very short time, such as a few microseconds, to perform this operation. So software zeroing operation requires much less time compared to hardware zeroing, so the user experience using these two zeroing methods is very different. Most users use software to reset sensors, which is more flexible and provides better real-time performance.
Kunwei Technology’s six axis force sensor currently does not provide a hardware reset command by default, unless the user deems it necessary to reset through the sensor firmware. We also recommend that users frequently perform software reset operations when using sensors, because the zero position state of the sensor is less affected by factors such as temperature in a short period of time, and the sensor data is always very accurate based on this zero position state. And the user is the operator of the entire actuator, very clear about the working status of the actuator, and can accurately specify a certain execution position as the zero position of the sensor. If users can frequently perform reset operations instead of only resetting once when powered on, the zero position of the sensor will not be affected by complex factors such as long-term temperature changes or changes in the status of the actuator, and the force information obtained by the user will always be accurate.

Search

Subscribe to our newsletter