A Multi-sensor Data Acquisition System Based on ADSP-2188M

Figure 1 Block diagram of multi-sensor data acquisition system

Figure 2 Ultrasonic receiving circuit

Figure 3 System software flow chart

System hardware design

The structure of the whole multi-sensor data acquisition system is shown in Figure 1. As can be seen from the figure, the system consists of DSP main controller, ultrasonic environment detection circuit, infrared sensor data acquisition circuit, PSD data acquisition circuit and communication module. The core of the system is ADSP 2188M , which mainly completes the control of various sensors, signal transmission and reception, information fusion and communication with the host PC host. Since the system has centralized preprocessing of sensor information, the amount of data for communication with the PC host will be greatly reduced, and the RS232 serial communication method can already meet the system's real-time requirements.

Ultrasonic environment detection circuit

The ultrasonic detection circuit is mainly composed of 16- channel ultrasonic generating circuits and ultrasonic receiving circuits. Mobile intelligent systems need to understand environmental information in real time during movement. Ultrasonic sensors are often evenly arranged around the system according to actual needs to form an environmental detection system. The principle of ultrasonic ranging is relatively simple, and the transit time method is generally used, that is:

D=Ct / 2 (1)

Where D is the distance between the mobile robot and the obstacle being measured, C is the transmission rate of the sound wave in the medium, t is the transit time, and the transmission rate of the sound wave in the air is:

(2)

Where T is the absolute temperature. In the case where high ranging accuracy is not required, C can generally be considered a constant. Measure the time interval t from the emission to the return of the ultrasonic wave , and then calculate the distance according to formula (1) .

The ultrasonic transmission part is mainly composed of the excitation signal generation circuit, buffer boost amplifier and ultrasonic transducer. The ultrasonic transmission process is: first, the DSP controls the multi-channel analog switch ADG438F to selectively start 4 of the 16 transmission circuits . Then, an I/O port of ADSP2188M generates a 40kHz modulated pulse wave with a pulse width of 25 μs , which passes through the buffer amplifier 7406 and the transformer boost amplifier circuit to generate an instantaneous high-energy excitation signal with an amplitude of 60V , and at the same time excites the 4 ultrasonic transducers to generate ultrasonic signals. The maximum opening time of ADG438F is 250ns , which will not affect the real-time performance of the system.

The reception and transmission of ultrasonic waves must work in a coordinated manner to ensure accurate and sensitive signal reception. The receiving part is mainly composed of receiving transducers, amplification filtering, shaping trigger output and other links, as shown in Figure 2. As the energy of ultrasonic waves decreases with the increase of transmission distance during propagation, the echo signal reflected from distant obstacles is generally weak ( mv level ) , so it needs to undergo multi-stage signal amplification processing. This system uses a three-stage amplification process to amplify the signal by about 1.5 million times, and then output it through the shaping circuit. The 4- way ultrasonic sensor receiving circuit uses the shaped signal as an interrupt request signal, and then connects it to the external interrupt pin of ADSP-2188M through a multi-way analog switch to trigger the external interrupt of DSP .

Experimental measurements show that the ultrasonic sensor has a sensitive range of approximately 30 degrees, a ranging range of 0.30 to 3 meters , and a repeated ranging accuracy of less than 1 %.

Infrared sensor data acquisition circuit

The infrared sensor data acquisition module is mainly composed of 8 -way infrared sensors, photoelectric isolation circuits and electronic switches. The detection distance of infrared sensors is generally short, and they are usually used to identify close-range obstacle targets, which can make up for the shortcomings of the blind spot characteristics of ultrasonic sensors to a certain extent. Its output is a switch value of 0 or 1. The working current of infrared sensors is generally about 25mA in static state, and can reach 60mA in dynamic operation . Therefore, the use of infrared sensor arrays will bring a great burden to the power supply system of the system. For this reason, a triode electronic switch is introduced into the infrared sensor data acquisition module to control the power supply of the infrared sensor. When the infrared sensor information needs to be collected, the power supply of the infrared sensor is turned on. After experimental measurement, the reaction speed of the electronic switch reaches 5.6 μs , which has almost no effect on the real-time performance of the system.

The data of 8 infrared sensors are directly connected to the high 8- bit data line of ADSP2188M through a photoelectric coupling circuit to achieve parallel acquisition.

PSD data acquisition circuit

The PSD data acquisition circuit consists of 4 PSD sensors , which are used to detect the target distance. Its working principle is to measure the distance of the reflecting object by measuring the phase difference between the emitted and reflected infrared light. The color of the measured object will not affect its ranging accuracy. PSD can make up for the shortcomings of ultrasonic sensors such as poor directionality and long measurement response time. We use Sharp 's GP2Y0A02YK infrared ranging device. This device can convert the change of phase difference into the change of output voltage, and can read the distance continuously. The measurement range is 20~150cm , and the output voltage range is 0~2.7V .

Software Description of the System

The system software is mainly composed of the main loop module, ultrasonic echo interrupt receiving module, communication module, etc. The specific flow chart is shown in Figure 3 .

Finally , no matter whether the echo is received or not, an appropriate delay should be performed to avoid the residual effect of ultrasound in indoor working environment .

Experimental Results

First, we tested the ranging performance of the ultrasonic sensor and PSD in the multi-sensor data acquisition system : different sensors of the system were used to measure the distance of the same target. The actual results of the ultrasonic sensor are shown in Table 1. The test shows that the repeated measurement accuracy of the two sensors is within 1 %, and the error between them is no more than 2 %, which can fully meet the requirements of general mobile intelligent systems.

At the same time, in order to verify the real-time performance of the entire multi-sensor data acquisition system, we calculated that the time taken by the main loop module of the system under the worst conditions ( that is, all ultrasonic sensors did not detect the echo signal ) was about 35.33ms . In other words, the update frequency of the system data can reach about 30Hz , proving that the system has good real-time performance.

Conclusion

This paper studies the hardware and software implementation and working principle of a high-performance, low-cost, low-power multi-sensor data acquisition system. The system uses the high-performance digital signal processor ADSP2188M as the core processor to centrally realize the acquisition and transmission of multiple infrared, ultrasonic and PSD sensor data. The experiment verifies the reliability, real-time nature of the hardware system and the effectiveness of the algorithm.