Views: 0 Author: Site Editor Publish Time: 2020-08-03 Origin: Site
Analysis of interference factors and measures are taken
One of the biggest disadvantages of using multiple sensors to transmit ultrasonic waves in parallel is that the interference is more serious, especially the signal interference between the sensors. The main factors that cause interference are as follows:
1) Installation error of the arduino ultrasonic sensors: The generation of ultrasonic waves is the mechanical vibration of the piezoelectric crystal, and the connection between the transmitting and receiving sensors is easy to cause interference; if the sensor and the ground are inclined or installed too low, the receiving sensor is easy to receive The reflected wave from the ground triggers the MCU to interrupt.
2) The influence of ultrasonic sidelobe: After the transmission wave is over, the first wave received by the receiving sensor is the crosstalk through wave, which is the beam sidelobe of the near source or directly reaches the receiving transducer through diffraction by the transmitting transducer Caused by the device [3]. Therefore, when installing the ultrasonic sensor, the distance between the two probes should be greater than 3cm.
3) Ultrasonic residual vibration interference: The transmitter sensor emits 8 sets of ultrasonic waves each time, each with 5~8 waveforms. When the obstacle is relatively close, the first set of waveforms can trigger the MCU interrupt. In this case, the ultrasonic wave that may be emitted when exiting the interrupt has not been completely attenuated. When the interrupt is turned on the next time, the MCU interrupt will be triggered immediately, resulting in interference data.
4) Ultrasonic cross interference: Multi-channel sensors transmit in parallel, the reflected ultrasonic waves received by the receiving sensor may not be emitted by the corresponding transmitting sensor, and the signals between the sensors are not synchronized, so that it is easy to cause the measurement time to be inaccurate. A lot of interference data appeared in the experiment is caused by this reason.
In order to shield the after-vibration and cross-interference of the ultrasonic distance sensors, the single-chip microcomputer adopts the low-level trigger interrupt mode, and the ultrasonic transmission is stopped in the interrupt service subroutine. After the MCU triggers the interrupt, during the period when the receiving sensor can receive the reflected wave, it has been cyclically executed in the interrupt service subroutine, waiting for the reflected ultrasonic wave to attenuate until it cannot be recognized by the system before exiting the interrupt.
Experimental calibration
The blind zone of the ranging system is 10cm. Since the obstacle avoidance measures are taken when the robot is set to be 40~50cm away from the obstacle in the program, the blind zone of the ranging system will not affect the obstacle avoidance of the robot. Install the ranging system on the robot, and use a plastic rod with a radius of 1cm to move in front of the robot to detect the sensitivity of the ranging system point by point. The detection point is selected on a straight line parallel to the ultrasonic sensor, with the centerline of the two sensors as the center, a point is selected every 5cm to both sides, and 4 points are selected on each side. It can be seen from the measurement results that the measurement error of the left road and the middle road is within 2%, and the error of the right road is too large. This difference is related to the installation accuracy of the sensor. In addition, the performance of the sensor can also cause this difference. In addition, the measured reference value 40 is obtained by visual inspection, and this measurement error will also affect the error analysis of the measurement result.
According to the above measurement method, the sensitivity of the ranging system is detected point by point, and the measurement range of the ultrasonic depth sensor
can be obtained. From the measurement range of the experimental calibration, a small area of the ranging system is not detected. This is mainly affected by the beam angle of the sensor. For targets that are not perpendicular to the transmitting beam, the sensor with a large beam angle can obtain stronger echoes. Signal, and the narrower the beam angle, the more beneficial it is to reduce the interference of scattered waves. It is very necessary to select a sensor with a suitable beam angle for omni-directional measurement. The experimental results show that within the safe distance of the robot, the ranging system can detect the environmental conditions in front of it in all directions and accurately, and the measurement data will not interfere with the obstacle avoidance needs of the robot.
4 Conclusion
In this paper, a high-performance robot ranging system is designed, which uses multiple sensors to work in parallel, which improves the real-time performance of ranging, and effectively shields system interference to meet the requirements of mobile robot avoidance. If the system is improved, it can be designed as a car reversing radar to improve the safety performance of the car.
In order to solve the deficiencies in ultrasonic positioning and navigation, this article provides a solution for the ultrasonic sensor-MB1004. This sensor is a proximity sensor with high and low level alarm signal output. The measurable range can reach 213cm, which is suitable for pedestrian detection and parking. Detection etc. When a pedestrian enters the detection range, MB1004 will output an alarm signal from low level to high level. At the same time, it also has the function of outputting the specific distance of the target, and outputting the distance data through RS232. MB1004 is a very low-cost ultrasonic sensor for human detection. It is also suitable for proximity area detection, pedestrian detection, booths/kiosks, automatic robot navigation, autonomous navigation, multi-sensor arrays, close-range detection and other fields.
