Multi-sensor fusion anti-collision system for microjets and control method of multi-sensor fusion anti-collision system

Abstract

The invention belongs to the field of water-guided laser safety control, and discloses a multi-sensor fusion anti-collision system for microjet and a control method thereof, and the anti-collision system comprises a linear sensor network, an analog quantity acquisition module, a switch, a data processing module and an execution module. Through multi-sensor data fusion, dynamic threshold judgment and setting of a corresponding grading response strategy, comprehensive monitoring and protection of the operation state of the micro-jet equipment are achieved, the detection precision of the sensors is greatly improved, tiny displacement changes can be captured in time, and the collision response time is remarkably shortened; a pre-trained risk assessment model is combined with historical processing data to dynamically optimize a dynamic threshold, so that the false alarm rate is effectively reduced; furthermore, by adopting a redundancy detection array formed by linear sensors, the non-dead-corner continuous monitoring of the micro-jet is realized, and the collision source is accurately positioned, so that the safety of the micro-jet equipment is improved, the processing quality is remarkably improved, and the service life of the equipment is remarkably prolonged.

Description

Technical Field

[0001] This invention belongs to the field of water-guided laser safety control technology, specifically relating to a collision avoidance system and control method for multi-sensor fusion of microjets. Background Technology

[0002] Water-guided lasers are a precision machining technology that combines high-energy lasers with micro-water jets. They are primarily used for high-precision cutting, drilling, and micromachining of semiconductors, medical devices, and brittle materials (such as glass and ceramics). Because water-guided laser equipment involves laser-water coupling, precision motion control, and complex machining environments, the design of its anti-collision system faces unique challenges.

[0003] Existing collision avoidance methods mostly employ sensor-based real-time collision avoidance monitoring or machine vision-assisted collision avoidance monitoring. However, sensor-based real-time collision avoidance monitoring is easily affected by water mist and processing debris during actual monitoring, leading to delayed warnings. Furthermore, it is limited by the reaction time of the sensor system and changes in the collision location, thus having certain limitations in practical applications. Existing machine vision-assisted collision avoidance monitoring methods utilize high-speed industrial cameras combined with image processing algorithms to monitor the morphology of the water jet and identify surface defects of the processed workpiece in real time to achieve collision avoidance monitoring. However, in practical applications, this machine vision-assisted collision avoidance monitoring method is significantly affected by water mist and moisture, easily leading to substantial influence on the monitoring results.

[0004] Existing collision avoidance methods suffer from long response times and high false trigger rates. Summary of the Invention

[0005] To address the aforementioned problems in the prior art, this invention provides a multi-sensor fusion anti-collision system and its control method for microjets. The technical problem to be solved by this invention is achieved through the following technical solution: In a first aspect, the present invention provides a collision avoidance system for multi-sensor fusion of microjets, wherein the microjets include: a shroud, an automated backplane, and a jet head, and the collision avoidance system includes: Linear sensor network, analog acquisition module, switch, data processing module, and execution module; among which, The linear sensor network is used to collect the minute errors of the fairing and the deformation displacement of the automated backplate, respectively. The analog quantity acquisition module is used to perform analog-to-digital conversion on the minute error and the deformation displacement, and output the converted minute error and the converted deformation displacement. The switch is used to send the converted minute error and the converted deformation displacement to the data processing module. The data processing module is used to acquire a dynamic threshold, and to determine the converted minute error and the converted deformation displacement based on the dynamic threshold, and output the corresponding control command; the dynamic threshold is obtained based on a pre-trained risk assessment model. The execution module is used to execute the corresponding hierarchical response strategy according to the control commands it receives, so as to avoid collisions.

[0006] Secondly, the present invention provides a collision avoidance control method for multi-sensor fusion of microjets, wherein the microjets include: a shroud, an automatic backplane, and a jet head, and the collision avoidance control method includes: The minute errors of the fairing and the deformation displacement of the automatic backplate were collected separately. The minute error and the deformation displacement are converted from analog to digital respectively, and the converted minute error and the converted deformation displacement are output. The converted minute errors and the converted deformation displacement are sent; The converted minute error and the converted deformation displacement are judged according to the dynamic threshold, and the corresponding control command is output; the dynamic threshold is obtained based on the pre-trained risk assessment model. Based on the control commands, execute the corresponding hierarchical response strategy to avoid collisions.

[0007] The beneficial effects of this invention are: The solution provided by this invention achieves comprehensive monitoring and protection of the micro-jet equipment's operating status through multi-sensor data fusion, dynamic threshold determination, and the setting of corresponding hierarchical response strategies. This greatly improves the detection accuracy of the sensors, enabling timely capture of minute displacement changes and significantly shortening the collision response time. The pre-trained risk assessment model dynamically optimizes the dynamic threshold based on historical processing data, effectively reducing the false alarm rate. Furthermore, by employing a redundant detection array composed of linear sensors, continuous monitoring of the micro-jet without blind spots is achieved, accurately locating the collision source, thereby improving the safety of the micro-jet equipment and significantly enhancing processing quality and equipment lifespan. Attached Figure Description

[0008] Figure 1 This is a topology diagram of a multi-sensor fusion anti-collision system for microjets provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the linear sensor detection and sensing part in a multi-sensor fusion anti-collision system for microjets provided in an embodiment of the present invention. Figure 3 This is a schematic diagram illustrating the steps of a multi-sensor fusion anti-collision control method for microjets provided in an embodiment of the present invention. Detailed Implementation

[0009] The present invention will be further described in detail below with reference to specific embodiments, but the implementation of the present invention is not limited thereto.

[0010] This invention provides a collision avoidance system and control method for multi-sensor fusion of microjets.

[0011] Firstly, we will introduce a collision avoidance system for multi-sensor fusion of microjets provided in the embodiments of the present invention.

[0012] This invention provides a multi-sensor fusion anti-collision system for microjets, wherein the microjets may include: a shroud, an automated backplane, and a jet head, such as... Figure 1 As shown, the collision avoidance system may include: Linear sensor network, analog acquisition module, switch, data processing module, and execution module; among which, A linear sensor network is used to collect the minute errors of the fairing and the deformation displacement of the automated backplate, respectively. The analog quantity acquisition module is used to perform analog-to-digital conversion on the minute error and the deformation displacement, and output the converted minute error and the converted deformation displacement. The switch is used to send the converted minute error and the converted deformation displacement to the data processing module; The data processing module is used to obtain dynamic thresholds and, based on the dynamic thresholds, to determine the converted minute errors and the converted deformation displacement, and output corresponding control commands; the dynamic thresholds are obtained based on a pre-trained risk assessment model. The execution module is used to execute the corresponding hierarchical response strategy based on the control commands it receives, in order to avoid collisions.

[0013] This invention aims to construct an active collision avoidance system using multi-sensor fusion technology and intelligent algorithms to address the problems of slow response speed, weak anti-interference capability, and insufficient predictive ability in existing technologies. The system achieves continuous, blind-spot-free monitoring of the jet head through a linear sensor array, and, combined with dynamic threshold determination and self-learning algorithms, significantly improves the system's reliability and adaptability, providing strong protection for the safe operation of water-guided laser equipment.

[0014] For ease of understanding, the various modules of the anti-collision system for multi-sensor fusion of microjets proposed in the embodiments of the present invention will be described below.

[0015] Linear sensor networks A linear sensor network may include: a first linear sensor and a second linear sensor; wherein, The first linear sensor is located at the jet end of the jet head; The second linear sensor is mounted on the automated backplane; The first linear sensor and the second linear sensor form a redundant detection array.

[0016] The linear sensor network collects the minute errors of the fairing and the deformation displacement of the automated backplate, and may include: The linear sensor network utilizes a first linear sensor to collect minute errors in the fairing in real time, and a second linear sensor to collect the deformation displacement of the automated backplate in real time; wherein, The fairing adopts an enclosed design.

[0017] This layout enables comprehensive monitoring of the displacement changes of the shroud and the automated backplate. The first linear sensor, combined with a high-precision linear sensor, detects minute error changes in the shroud, thus actively monitoring and protecting the jet head throughout its entire lifecycle. The second linear sensor detects the deformation and displacement of the automated backplate, ensuring safe monitoring of the overall equipment structure. The sensor detection accuracy reaches 0.001mm, and it can automatically collect initial data at the moment of equipment startup.

[0018] A schematic diagram of the sensing part of a linear sensor, as shown below. Figure 2 As shown, the first linear sensor is connected to the shroud via a connecting rod. The installation angle of the first linear sensor can be 45°, which is the angle between the first linear sensor and the connecting rod. The diameter of the connecting rod can be 2mm, the diameter of the jet head can be 30mm, the diameter of the maximum cross-section of the shroud can be 50mm, and the distance between the shroud and the jet head can be 2mm.

[0019] The shroud features a wraparound design for the jet head. Signal acquisition is initiated when any servo axis of the data processing module moves in any direction. This wraparound design maximizes protection of the jet head against collisions in the horizontal 360° and vertical directions. Furthermore, the shroud design was optimized through multiple simulations and on-machine testing across various collision scenarios. Ensuring safety and feasibility, the final design gap between the shroud and the jet head was set to 2mm, maximizing the slender characteristics of the jet head and guaranteeing the reliable operation of the anti-collision system. Regarding signal detection, the sensor's minimum resolution is 0.001mm, with a response period of 10ms. Under rapid movement, the braking distance of the equipment is 0.003mm, and under normal processing conditions, the braking distance can be stably maintained within 0.005mm.

[0020] The second linear sensor can provide redundant protection for the anti-collision system. It uses single-sensor data measurement and detects the elastic displacement of the automatic backplate in the equipment during the operation of any servo axis. The second linear sensor automatically collects data at the moment of equipment startup.

[0021] This invention overturns the traditional point / area detection method by using a continuously linearly arranged sensor array (linear sensor) to achieve 360° continuous monitoring of the jet head without blind spots, so as to determine the precise coordinates of the collision.

[0022] Analog signal acquisition module The analog quantity acquisition module performs analog-to-digital conversion on the minute error and the deformation displacement, and outputs the converted minute error and the converted deformation displacement.

[0023] Optionally, the analog acquisition module can select a corresponding analog-to-digital converter to perform analog-to-digital conversion on minute errors and deformation displacements.

[0024] switch The switch sends the converted minute error and the converted deformation displacement to the data processing module.

[0025] Data processing module The data processing module determines the converted minute error and the converted deformation displacement based on the dynamic threshold, and outputs the corresponding control commands.

[0026] Optionally, the data processing module can be a Siemens S7-200 processor, configured with a multi-threaded real-time processing architecture, which can quickly process sensor data and generate control commands.

[0027] The data processing module determines the converted minute error and the converted deformation displacement based on dynamic thresholds, and outputs corresponding control commands, including: When the difference between two consecutive small errors is greater than or equal to the dynamic threshold, and the difference between two consecutive deformation displacements is less than the dynamic threshold, the first braking command is output. When the difference in deformation displacement after two consecutive conversions is greater than or equal to the dynamic threshold, and the difference in minute error after two consecutive conversions is less than the dynamic threshold, the second braking command is output. When the difference between the small errors after two consecutive conversions is greater than or equal to the dynamic threshold, and the difference between the deformation displacement after two consecutive conversions is greater than or equal to the dynamic threshold, the third braking command is output.

[0028] Specifically, the processing procedure for the small error after conversion corresponding to the first linear sensor is as follows: The system detects the start of operation of any servo axis, records the small error 1D1 after the current conversion, records the small error 1D2 after the real-time conversion during operation, obtains the difference 1D1-1D2 between two adjacent small errors, compares 1D1-1D2 with the dynamic threshold, and outputs the corresponding control command.

[0029] For the converted deformation displacement corresponding to the second linear sensor, the processing procedure is as follows: The system detects the start of operation of any servo axis, records the deformation displacement 2D1 after the current conversion, records the deformation displacement 2D2 after the real-time conversion during operation, obtains the difference 2D1-2D2 between two adjacent conversions, and compares 2D1-2D2 with the dynamic threshold to output the corresponding control command.

[0030] Understandably, the linear sensor in this embodiment of the invention can detect the change in displacement of the jet head in real time, with a detection accuracy of 0.001mm. When the difference between the first linear sensor or the linear sensor reaches the dynamic threshold, the anti-collision system is activated, the laser is instantly cut off, all servo axes corresponding to the linkage data processing module stop running, and the control command corresponding to the emergency stop is output. The specific collision response time can be shortened to less than 8ms, which is far less than the collision response time of 50ms of the traditional solution.

[0031] The dynamic threshold is obtained based on a pre-trained risk assessment model; for example, the obtained dynamic threshold could be 0.003 mm. The pre-trained risk assessment model is trained on test datasets of several collision scenarios. This model can dynamically optimize the threshold curve by incorporating historical processing data, thereby significantly reducing the false alarm rate and improving the reliability of the collision avoidance system. The test dataset can include data from over 100 collision scenarios, meeting the needs of full-scenario path applications.

[0032] For example, when the dynamic threshold is 0.003 mm: When 1D1-1D2 is greater than or equal to 0.003mm and 2D1-2D2 is less than 0.003mm, the first braking command is output; When 2D1-2D2 is greater than or equal to 0.003mm and 1D1-1D2 is less than 0.003mm, the second braking command is output. When 1D1-1D2 is greater than or equal to 0.003mm and 2D1-2D2 is greater than or equal to 0.003mm, the third braking command is output.

[0033] Execution module The execution module may include: Safety module, audible and visual alarm module, and motor braking module.

[0034] The execution module executes the corresponding hierarchical response strategy based on the control commands it receives to avoid collisions, which may include: When the control command is the first braking command, the safety module controls the motor braking module to brake urgently to stop the micro-jet from running, and the first linear sensor alarm signal is triggered by the audible and visual alarm module. When the control command is the second braking command, the safety module controls the motor braking module to brake urgently to stop the micro-jet from running, and the second linear sensor alarm signal is triggered by the audible and visual alarm module. When the control command is the third braking command, the safety module controls the motor braking module to brake urgently to stop the microjet from running, and the first linear sensor alarm signal and the second linear sensor alarm signal are triggered by the audible and visual alarm module.

[0035] In practical applications, the anti-collision system proposed in this invention can significantly improve the safety and processing quality of water-guided laser equipment. For example, in the high-precision cutting of semiconductor materials, microjets need to move at high speeds on the surface of complex workpieces. Traditional anti-collision systems struggle to identify and address potential issues such as uneven surfaces or loose clamps on the workpiece surface in a timely manner. However, this invention utilizes a linear sensor network to achieve continuous, blind-spot-free monitoring of the jet head within the microjets, thereby accurately locating the coordinates of any collision. Understandably, the anti-collision system proposed in this invention not only improves the safety of water-guided laser equipment but also significantly enhances processing quality and equipment lifespan, demonstrating significant industrial application value.

[0036] Secondly, corresponding to the above-described anti-collision system embodiments, this invention also provides an anti-collision control method for multi-sensor fusion of micro-jets, such as... Figure 3 As shown, the microjet can include: a shroud, an automatic backplate, and a jet head; the anti-collision control method can include: S1, collects the minute error of the fairing and the deformation displacement of the automatic backplate respectively; S2, perform analog-to-digital conversion on the small error and the deformation displacement respectively, and output the converted small error and the converted deformation displacement; S3 sends the converted minute error and the converted deformation displacement; S4, based on the dynamic threshold, determine the converted minute error and the converted deformation displacement respectively, and output the corresponding control command; the dynamic threshold is obtained based on the pre-trained risk assessment model; S5 executes the corresponding hierarchical response strategy according to the control command to avoid collisions.

[0037] Understandably, in the anti-collision control method proposed in the embodiments of the present invention, the specific execution process corresponding to step S1 is referred to the working principle of the linear sensor network in the anti-collision system proposed in the first aspect; the specific execution process corresponding to step S2 is referred to the working principle of the analog quantity acquisition module in the anti-collision system proposed in the first aspect; the specific execution process corresponding to step S3 is referred to the working principle of the switch in the anti-collision system proposed in the first aspect; the specific execution process corresponding to step S4 is referred to the working principle of the data processing module in the anti-collision system proposed in the first aspect; and the specific execution process corresponding to step S5 is referred to the working principle of the execution module in the anti-collision system proposed in the first aspect. Further details will not be elaborated here.

[0038] This invention, through multi-sensor data fusion, dynamic threshold determination, and the setting of corresponding hierarchical response strategies, achieves comprehensive monitoring and protection of the micro-jet equipment's operating status, greatly improving the sensor's detection accuracy, enabling timely capture of minute displacement changes, and significantly shortening the collision response time. The pre-trained risk assessment model, combined with historical processing data, dynamically optimizes the dynamic threshold, effectively reducing the false alarm rate. Furthermore, by employing a redundant detection array composed of linear sensors, continuous monitoring of the micro-jet without blind spots is achieved, accurately locating the collision source, thereby improving the safety of the micro-jet equipment and significantly enhancing processing quality and equipment lifespan.

[0039] It should be noted that, in the description of this invention, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0040] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the method embodiments are basically similar to the system embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the system embodiments.

[0041] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. A collision avoidance system for multi-sensor fusion of microjets, the microjets comprising: The shroud, automatic backplate, and jet head are characterized in that the anti-collision system comprises: Linear sensor network, analog acquisition module, switch, data processing module, and execution module; among which, The linear sensor network is used to collect the minute errors of the fairing and the deformation displacement of the automated backplate, respectively. The analog quantity acquisition module is used to perform analog-to-digital conversion on the minute error and the deformation displacement, and output the converted minute error and the converted deformation displacement. The switch is used to send the converted minute error and the converted deformation displacement to the data processing module. The data processing module is used to acquire a dynamic threshold, and to determine the converted minute error and the converted deformation displacement based on the dynamic threshold, and output the corresponding control command; the dynamic threshold is obtained based on a pre-trained risk assessment model. The execution module is used to execute the corresponding hierarchical response strategy according to the control commands it receives, so as to avoid collisions.

2. The anti-collision system for multi-sensor fusion of microjets according to claim 1, characterized in that, The linear sensor network includes: a first linear sensor and a second linear sensor; wherein, The first linear sensor is disposed at the jet end of the jet head; The second linear sensor is mounted on the automated backplane; The first linear sensor and the second linear sensor constitute a redundant detection array.

3. The anti-collision system for multi-sensor fusion of microjets according to claim 2, characterized in that, The linear sensor network collects minute errors in the fairing and deformation displacement of the automated backplate, including: The linear sensor network utilizes a first linear sensor to collect minute errors in the shroud in real time, and a second linear sensor to collect the deformation displacement of the automated backplate in real time; wherein, The fairing is designed in a wraparound manner.

4. The anti-collision system for multi-sensor fusion of microjets according to claim 1, characterized in that, The data processing module determines the converted minute error and the converted deformation displacement based on dynamic thresholds, and outputs corresponding control commands, including: When the difference between two consecutive small errors is greater than or equal to the dynamic threshold, and the difference between two consecutive deformation displacements is less than the dynamic threshold, the first braking command is output. When the difference between the deformation displacements after two consecutive conversions is greater than or equal to the dynamic threshold, and the difference between the small errors after two consecutive conversions is less than the dynamic threshold, the second braking command is output. When the difference between the small errors after two consecutive conversions is greater than or equal to the dynamic threshold, and the difference between the deformation displacement after two consecutive conversions is greater than or equal to the dynamic threshold, the third braking command is output.

5. The anti-collision system for multi-sensor fusion of microjets according to claim 1, characterized in that, The pre-trained risk assessment model was trained based on test datasets of several collision scenarios.

6. The anti-collision system for multi-sensor fusion of microjets according to claim 1, characterized in that, The dynamic threshold is 0.003 mm.

7. A collision avoidance system for multi-sensor fusion of microjets according to claim 4, characterized in that, The execution module includes: Safety module, audible and visual alarm module, and motor braking module.

8. A collision avoidance system for multi-sensor fusion of microjets according to claim 7, characterized in that, The execution module executes a corresponding hierarchical response strategy based on the control commands it receives to avoid collisions, including: When the control command is the first braking command, the safety module controls the motor braking module to brake urgently to stop the micro-jet from running, and the first linear sensor alarm signal is triggered by the audible and visual alarm module. When the control command is the second braking command, the safety module controls the motor braking module to brake urgently to stop the micro-jet from running, and the second linear sensor alarm signal is triggered by the audible and visual alarm module. When the control command is the third braking command, the safety module controls the motor braking module to brake urgently to stop the microjet from running, and the first linear sensor alarm signal and the second linear sensor alarm signal are triggered by the audible and visual alarm module.

9. A collision avoidance control method for multi-sensor fusion of microjets, wherein the microjets include: The shroud, automatic backplate, and jet head are characterized in that the anti-collision control method includes: The minute errors of the fairing and the deformation displacement of the automatic backplate were collected separately. The minute error and the deformation displacement are converted from analog to digital respectively, and the converted minute error and the converted deformation displacement are output. The converted minute errors and the converted deformation displacement are sent; The converted minute error and the converted deformation displacement are judged according to the dynamic threshold, and the corresponding control command is output; the dynamic threshold is obtained based on the pre-trained risk assessment model. Based on the control commands, execute the corresponding hierarchical response strategy to avoid collisions.

Images