Design and Optimization of Obstacle Avoidance Algorithm for ultrasonic sensors of Industrial robots
The development of industrial robots has been nearly 70 years. Under the background of artificial intelligence and computer technology, the advantages of intelligence and ease of use are prominent. The independent development and man-machine cooperation are becoming one of the development trends in the field of industrial robots. With the technological innovation in the field of logistics, electronics and industrial production, and the conversion of old and new kinetic energy, the realization of efficient and high-precision obstacle avoidance when the industrial robot moves is the key to the transformation and upgrading of the industrial robot industry and the industrial application. Its obstacle avoidance performance is also an indicator to measure the practicability of industrial robots. At present, obstacle avoidance algorithms represented by fuzzy logic control and artificial potential field method play a crucial role in the operation of industrial robots. However, this kind of technology is only suitable for obstacle avoidance in simple environment, and the obstacle avoidance effect is not good for multi-obstacle environment and complex working conditions, and there are common problems such as pause and swing in the middle. In contrast, the ultrasonic sensor method can provide a wide field of vision and obstacle information for the robot in obstacle avoidance, and provide a new research idea and direction for the design of obstacle avoidance strategy for industrial robots.
1, the principle of ultrasonic sensor obstacle avoidance
Ultrasonic sensor is composed of piezoelectric wafer, is based on the ultrasonic wave length, high frequency, propagation orientation, low diffraction frequency characteristics of the design of the sensor, because of the large electromechanical coupling coefficient, output energy is large, has a high sensitivity in the work, can be suitable for high load conditions of the sensor operation needs. The application of ultrasonic technology based on longitudinal oscillation can propagate in different environments of solid, liquid and gas, and there are obvious differences in the propagation efficiency. However, no matter it is high frequency transmission in liquid and solid, or low frequency transmission in air, it is accompanied by attenuation phenomenon. In view of this characteristic of ultrasonic, in the process of industrial robots, Ultrasonic ranging module can be used to measure the distance between it and obstacles in real time. In the set safe range, the robot can automatically avoid obstacles. The obstacle avoidance method is evolved and formed in practice on the basis of time difference ranging method, that is, according to the movement time and distance of ultrasonic wave in the air within a unit time, the ultrasonic wave in the air from the signal sent by the transmitter to the receiver received echo distance, the calculation formula is: S=Tυ
In the expression:
S is for the distance the ultrasound travels through the air; T is for the time it takes the ultrasonic wave to move; υ represents the travel distance of ultrasonic waves in the air in unit time.
The ultrasonic sensor in the industrial robot walking obstacle detection mode is continuous detection, referring to the common ultrasonic sensor module, the detection error is generally within 4%, so the use of ultrasonic sensor obstacle avoidance, can obtain good obstacle avoidance effect.
2. Obstacle Avoidance algorithm and process design of industrial robot ultrasonic sensor
Based on the principle of obstacle avoidance of ultrasonic sensors, in the design of obstacle avoidance of industrial robots, based on the relationship between sound speed, time and obstacle distance, the expression of the three can be expressed as follows: S=1/2ct
In the above expression, C represents the speed of sound;
S represents the distance between the industrial robot and the obstacle; T represents the time difference between the moment the pulse is transmitted and the moment the first echo arrives.
According to the above obstacle avoidance formula, ultrasonic ranging principle, using the MATLAB platform Line function, FIS fuzzy reasoning system editor, integration of industrial robot obstacle information acquisition, detection, processing, fuzzy reasoning process, to establish a simple model, reduce the modeling process on the robot's own mechanical model dependence. With the help of fuzzy reasoning and control, the accuracy of obstacle avoidance algorithm in application is strengthened, which is convenient for the robot to recognize and control the obstacles in all directions of left, right and front. Taking the designed ultrasonic obstacle avoidance algorithm as an example
when the industrial robot is working, firstly, according to the requirements of industrial production and functional positioning of the robot, the parameter initialization is set, and the obstacle avoidance parameters of the robot are debuted, so that the industrial robot can timely choose the barrier-free route within a safe distance. After that, ultrasonic sensors are used to collect acoustic signals around robots, detect the distance between industrial robots and obstacles, combine the fuzzy reasoning method, predict the progress of obstacles, recommend reliable routes according to the position of robots, generate detection results, and then transmit the data to the display screen configured by industrial robots. It can display obstacle information and distance detection results in different directions. Then, under the obstacle, the robot can avoid the obstacle, rotate 90° and continue to choose a safe travel. If the robot judges that there is no obstacle in the predetermined route, it can directly select the route corresponding to the orientation and complete the specified task.
3. Optimization strategy of obstacle avoidance algorithm for industrial robot ultrasonic sensor
Considering the basic ultrasonic sensor algorithm in short distance measurement accuracy gap problem, through the delay circuit time, to control the ultrasonic ranging accuracy in the process of obstacle avoidance, the optimized obstacle avoidance algorithm expression is: △t=s1t1-s2t2/s2-s1
s1 and s2 indicate the known measurement distance; △t represents the delay time; t1 and t2 indicate the corresponding measurement time.
According to the optimized ultrasonic obstacle avoidance algorithm, the distance parameters between industrial robots and obstacles are analyzed and set, including the front obstacle, the left obstacle and the right obstacle [16]. Ultrasonic sensors are used to detect the information of various parts. When the industrial robot is required to walk in front, when the distance between the obstacle and the industrial robot is more than 25cm, it is regarded as a safe route and the robot can continue to walk. As the distance from the obstacle is pulled in, when the obstacle distance is less than 25cm, the obstacle information on the left side and the right side of the industrial robot is detected. When the distance to the left side and the right side is identified and the machine person is more than 30cm, it can rotate 90° to the left or right side before moving. If the safe distance between the left and right sides is less than 25cm, the industrial robot can back up for a certain distance. When the distance of one side is more than 25cm, it can walk in this direction until the best barrier-free route is chosen.
4. Verification of obstacle avoidance algorithm for ultrasonic sensor
In order to verify the accuracy and reliability of the obstacle avoidance algorithm of ultrasonic sensor, AIR3 industrial robot produced by XX company and 200F18TR-2 ultrasonic sensor produced by XX Company were used as test tools. Among them, the industrial robot has 6 axes, with the maximum motion range of 560mm and positioning accuracy of ±0.02mm. The motion ranges of the 6 axes are: - 170 ° / + 170 °, 110 ° / + 120 °, 110 ° / + 155 °, 200 ° / + 200 °, 120 ° / + + 350 ° 120 °, 350 ° /, The velocities of the six axes are respectively 450°/s, 450°/s, 540°/s, 800°/s, 800°/s and 800°/s. The nominal frequency of ultrasonic sensor is 200.0±10%KHz, the measuring range is 0.07-0.8M, the standard test board is 100x100mm, and the output mode is NPN/PNP/0-10V/4-20mA. Based on the current working speed of the industrial robot, make it move in the direction of the obstacle, measure the minimum distance between the industrial robot and the obstacle, and test the operation state of the industrial robot under three different working conditions, namely, the obstacle in front, the obstacle on the left and the obstacle on the right:
After the running state test of industrial robots under different working conditions, it is found that the ultrasonic sensor obstacle avoidance algorithm has fixed requirements for the operating environment of industrial robots in obstacle avoidance. The robot walking process needs to ensure that the road surface is smooth and smooth, without pits and uneven phenomena, and it can be processed according to the set parameters for obstacle avoidance. Once there are pits on the road surface, The measurement accuracy of ultrasonic sensor will be affected, and the robot may deviate from the original route with a certain probability. In addition, within the range of the ultrasonic sensor, it has a good measurement effect for slightly distant obstacles, while there may be blind spots in the detection of near-distance obstacles. However, thanks to the application of fuzzy reasoning strategy in the test, the ultrasonic sensor can enhance its ability to identify obstacles, improve its response ability to deal with obstacles, and reduce detection errors. This also effectively guides the normal operation of industrial robots, but the accuracy of fuzzy inference strategy is limited for close range ultrasonic sensors.
5 Conclusion
In this paper, the use of ultrasonic sensor to design the industrial robot obstacle avoidance algorithm, establish the industrial robot obstacle avoidance process, determine the barrier to the industrial machine human safety distance control threshold, ranging sensitivity, high precision, can be applied to the general environment of industrial robot simple operation, for complex conditions, Especially the dynamic obstacle avoidance processing and close distance obstacle avoidance design, the ultrasonic sensor obstacle avoidance algorithm also has its own disadvantages, the ultrasonic obstacle avoidance algorithm and genetic algorithm, artificial intelligence technology combined with algorithm optimization, is an important proposition in the future of industrial robot ultrasonic obstacle avoidance.
References:
[1] Takei Y,Matsui R,Minh-Dung N, et al.Velocity measurement using MEMS ultrasonic sensor for non-invasive blood pressure measurement[J]. IEEJ Transactions on Sensors and Microma⁃ chines, 2018,138(2):54-58.
[2] AlMohimeed I,Ono y.Ulu trasound measurement of skeletal mus⁃ cle contractile parameters using flexible and wearable Single element ultrasonic sensor [J]. Journal of Sensors, 2020, 20 (13) : 3616.
[3] Hua Dan. Design of Industrial Robot Visual Detection and Obstacle avoidance System based on Blockchain technology [J]. Computer Measurement and Control,20,28(7):69-73.
[4] YU Yan. Development and Application of Ultrasonic Obstacle Avoidance Technology for Intelligent Robot [J]. Southern Agricultural Machinery,2019,50(21):263.
[5] Wu Chuanhui, Lu Duxiao, Luo Zhonghui, et al. Development and Application of Ultrasonic Obstacle Avoidance Technology for Intelligent Robot [J]. Machine Tool & Hydraulics,2019,47(9):46-50,57.
