topic:
Design and manufacture a seismic source orientation detection device. The detection area is a square with a side length of 500 mm. The X and Y coordinate systems are marked along the horizontal and vertical edges, and the O point is the origin of the coordinates. The sensor layout area is in the lower right corner, which is a square area with a side length of 100mm. When 10g, 5g, and 2g weights fall in the seismic source area, the detection device can display the coordinate position (X, Y) of the weight falling point in the coordinate system in the figure.
Preliminary study of the plan:
The reason for the failure to deduce the value of x and y through the weight recognition through the camera :
the camera cannot illuminate the entire board and there will be a blind area, adding two cameras will have an overlapping area and because the object is too small, it is difficult to distinguish the size of the weight.- The distance value is obtained through multiple ultrasonic ranging modules, and the x, y value is obtained through algorithm calculation.
Reason for failure :
The distance measuring module cannot measure the weight, and the distance of the board behind can be measured through the weight.
The reason for the failure of position estimation through the xyz acceleration value of mpu6050 :
The maximum data output rate of the serial port of mpu6050 is 100HZ, and the sampling frequency of mpu6050 through IIC is 1KHZ, and the sampling rate calculated by physical formula can not meet the requirements at all.
The final proposal:
Initial idea : through the level change of the digital value of the sound sensor, the intermediate time difference can be obtained through the TDOA positioning algorithm to obtain the x and y values.
The reason for the failure of the initial idea : through the display of the oscilloscope, since the module judges the high and low levels through the comparator, the comparison voltage can be changed through the adjustment of the sliding resistor, and it is difficult to achieve the consistency of the comparison level, resulting in the sensitivity of the module Different, the obtained sine wave signal period is different. (some when the trigger is reached and others still don't)
The final idea : The frequency and duration of the triggering moment can be obtained through the analog signal of the sound sensor. According to the same cycle of the signal received by the theoretical sensor, the phases of multiple sensors can be compared, and the phase difference between points can be obtained by comparing the phase shift. After the time difference, the x and y values can be obtained through the TDOA positioning algorithm.
Actual difficulties :
- The sound module detects the sound transmitted through the vibration of the board. Since the density of the board itself may not be exactly the same, the sound received by different modules at different locations may be inconsistent.
- The sound module receives sound through the microphone, and the degree of fit between the placement of the microphone and the board leads to different strengths of reception, which may even lead to errors in receiving human voices
- According to the circuit diagram of the module, the sound received by the microphone is not filtered, resulting in excessive noise, which can easily lead to misjudgment.
- Passing the received analog signal through the filter of another module will cause signal distortion between high and low voltages, resulting in a shape similar to a square wave.
- Because there is a sliding resistor that needs to be adjusted for sensitivity, the sensitivity between modules sometimes varies greatly, and it is difficult to compare the phases of multiple signals displayed by the oscilloscope.
Solution : Establish a cross-correlation function through the sinusoidal signals between multiple sensors to collect a large number of points, and perform function fitting to obtain the coordinates of x and y.