Mars detection robot and detection system for exhaust pipe of transport truck

Abstract

The invention belongs to the technical field of industrial safety monitoring and specialized robots, particularly relates to a Mars detection robot and system for an exhaust pipe of a transport truck, and solves the problems of vehicle type differentiation and complex space positioning by adopting accurate self-positioning guided by a database. The self-adaptive and high-precision alignment of exhaust ports of vehicle types with different sizes, different heights and different orientations is realized; on the basis, interference of strong background light such as sunlight is effectively isolated from the physical level through an original mechanical arm control light shield local light shielding method, and the reliability of detection signals is greatly improved; meanwhile, through further combination with a sensor self-cleaning module, optical components are actively kept clean, and long-term stable operation of the system is guaranteed.

Description

A robot and system for detecting sparks from the exhaust pipes of transport trucks Technical Field

[0001] This invention belongs to the field of industrial safety monitoring and special robot technology, specifically relating to a robot and system for detecting sparks from the exhaust pipes of transport trucks. Background Technology

[0002] In the production process of chemical products such as zirconium dioxide, ammonia gas is often present or leaked in the workshop environment due to process requirements. Combustible gases such as ammonia. When the concentration of ammonia in the air reaches the lower explosive limit (approximately 15%–28%), it can easily ignite or even explode if it encounters an open flame or high heat source, posing a serious threat to personnel and production facilities. Diesel and gas-powered trucks transporting raw materials or finished products frequently enter such workshops, and their exhaust pipes may emit unburned high-temperature carbon particles (commonly known as "sparks") when running or shutting down, becoming a typical mobile ignition source.

[0003] Currently, routine safety management measures at workshop entrances mainly rely on driver self-inspection or visual checks by security personnel to ensure vehicles are equipped with spark arresters (fireproof caps). However, this method has significant drawbacks: First, it is manual, inefficient, and unreliable, especially prone to oversights at night or during peak hours; second, even if spark arresters are installed, they may fail due to carbon buildup or damage after prolonged use, and it is difficult for humans to assess this in real time. Therefore, there is an urgent need for an automated, highly reliable technology to perform online real-time detection and risk assessment of sparks in vehicle exhaust before they enter hazardous areas. The inventors found that in the prior art, the detection of exhaust pipe sparks is mainly based on the photoelectric sensing principle, which can be roughly divided into two categories: (1) Fixed detection: Infrared or ultraviolet flame sensors are installed above or to the side of the workshop entrance passage to scan the exhaust pipe area of ​​passing vehicles. However, such systems have the following defects: severe environmental interference: workshops are usually semi-open or have a large area of ​​lighting. Strong and changing natural light (especially direct or reflected sunlight) will cause photoelectric sensors to saturate or generate background noise, which is very easy to cause false alarms (judging sunlight as sparks) or missed alarms (spark signals are drowned out by strong light).

[0004] Blind spots and fixed viewing angles: Vehicle models, exhaust pipe heights and orientations vary, and the installation angle and field of view of fixed sensors are limited, making it difficult to ensure effective observation of all vehicle exhaust outlets from the best angle without blind spots.

[0005] Sensor contamination and performance degradation: Vehicle exhaust contains a large amount of oil, soot and water vapor, which will continuously contaminate the optical window of the sensor. Existing systems lack effective online self-cleaning capabilities, causing the sensor sensitivity to drop sharply over time and making it difficult to maintain detection accuracy.

[0006] (2) Handheld inspection: The staff holds the instrument close to the exhaust pipe for inspection. Although this method is relatively flexible, it is extremely inefficient and cannot achieve automated continuous screening. In addition, it exposes the staff to high temperature, exhaust gas and potential risk environment, which does not conform to the development trend of modern industrial automation and safe human-machine isolation. Summary of the Invention

[0007] This invention addresses the technical problems existing in the prior art by providing a robot and system for detecting sparks from the exhaust pipes of transport trucks, effectively solving the problems present in the prior art.

[0008] To achieve the above objectives, the technical solution adopted by this invention is as follows: a robot for detecting sparks from the exhaust pipes of transport trucks, comprising an autonomous mobile platform, an integrated detection module, and a controller; the upper part of the autonomous mobile platform is provided with an embedded cavity for storing the integrated detection module, and the bottom of the embedded cavity is provided with a lifting platform; the integrated detection module is connected to the lifting platform via a multi-degree-of-freedom robotic arm; the integrated detection module includes a sensor module for spark detection and temperature detection, a light-shielding cover, and a sensor self-cleaning module; wherein, the integrated detection module transmits the collected sensor data to the controller; the controller performs the following processing procedure: in response to a vehicle arriving in the detection area. Based on the signal and vehicle identification results, and according to a pre-built vehicle-detection point database, the optimal detection and parking target position of the Mars detection robot, the offset vector of the exhaust port relative to the optimal detection and parking target position, and the recommended posture of the multi-degree-of-freedom robotic arm are determined. The optimal detection and parking target position is transmitted to the autonomous mobile platform. Based on the navigation and obstacle avoidance strategies built into the autonomous mobile platform, the Mars detection robot is controlled to move to the optimal detection and parking target position. Based on the offset vector and the recommended posture of the multi-degree-of-freedom robotic arm, the integrated detection module is controlled to reach the preset position of the exhaust port to perform data acquisition. Based on the acquired sensor data, the Mars detection result is obtained.

[0009] Furthermore, a telescopic door is provided at the top of the embedded cavity. The telescopic door is controlled by the controller, so that when the Mars detection robot moves to the optimal detection and parking target position, it opens and extends, and the multi-degree-of-freedom robotic arm and integrated detection module are pushed out of the embedded cavity through the telescopic rod connecting the bottom of the lifting platform and the bottom of the embedded cavity; and after the detection task is completed and the multi-degree-of-freedom robotic arm and integrated detection module are reset and put back into the embedded cavity, the telescopic door is closed by the controller.

[0010] Furthermore, the integrated detection module is equipped with a horn-shaped light shield, the outermost inner diameter of which is larger than the outer diameter of the exhaust port.

[0011] Furthermore, the multi-degree-of-freedom robotic arm adopts a two-stage robotic arm structure, and the sensor self-cleaning module of the integrated detection module includes a micro air tank and a micro air pump integrated module. The micro air tank and the micro air pump integrated module are connected to the air nozzle located at the sensor module position through a pipeline.

[0012] Furthermore, the integrated detection module is connected to the lifting platform via a multi-degree-of-freedom robotic arm, wherein an electrically rotating base is provided between the multi-degree-of-freedom robotic arm and the lifting platform.

[0013] Furthermore, the integrated detection module and the end of the multi-degree-of-freedom robotic arm are equipped with a gimbal, which allows for multi-directional tilt adjustment of the integrated detection module to adapt to different tilt angle designs of the exhaust port.

[0014] Furthermore, the sensor module includes a composite optical sensor unit and a temperature sensor unit, wherein the composite optical sensor unit includes a first sensor sensitive to the characteristic spectrum of Martian light and a second sensor responding to ambient white light.

[0015] Further, the process of obtaining the Mars detection result based on the collected sensor data specifically involves the following steps: acquiring real-time data from the first and second sensors; calculating the steady-state background values ​​corresponding to the two sensors based on the acquired sensor data; extracting the signal peak value based on the acquired steady-state background value; calculating the spectral feature ratio based on the acquired signal peak value; and determining whether it is a Mars based on the acquired signal peak value and the spectral feature ratio. Specifically, if the signal peak value of the first sensor is less than a first preset threshold, it is determined to be noise; if it is greater than the first preset threshold, it is determined whether the spectral feature ratio is greater than a second preset threshold. If it is, it is determined to be a Mars; if it is greater than a third threshold but less than or equal to the second threshold, it is determined to be a possible Mars and the next round of detection is performed; if it is less than or equal to the third threshold, it is determined not to be a Mars.

[0016] Furthermore, the integrated detection module performs data acquisition upon reaching the preset position of the exhaust port. Specifically, it acquires the optimal target position for robot detection and parking corresponding to the current vehicle model, and combines the offset vector of the exhaust port relative to the optimal target position, the unit vector of the exhaust port orientation, the recommended posture of the multi-degree-of-freedom robotic arm, and the safe distance between the sunshade and the exhaust port. Using the robot's parking position as a reference, it calculates the precise coordinates of the exhaust port in the world coordinate system by adding the offset vector. It controls the multi-degree-of-freedom robotic arm to move to the recommended posture to achieve coarse alignment between the integrated detection module and the exhaust port. Based on the calculated precise coordinates of the exhaust port in the world coordinate system, combined with the safe distance between the sunshade and the exhaust port and the unit vector of the exhaust port orientation, it calculates the target position of the sunshade at the end of the robotic arm. It controls the end of the robotic arm to move to the target position, and uses gimbal control to make the central axis of the sunshade coincide with the unit vector of the exhaust port orientation, performs sensor data acquisition, and processes the data based on the aforementioned spark detection process to obtain the spark detection result.

[0017] A spark detection system for exhaust pipes of transport trucks includes a spark detection robot for exhaust pipes of transport trucks, a monitoring module, and a remote control platform. The monitoring module is used to acquire images of the detection area in real time and transmit them to the remote control platform. The remote control platform receives the images of the detection area and determines the arrival status and vehicle type of the vehicle to be inspected based on a built-in image recognition algorithm. When the vehicle to be inspected arrives, it sends the vehicle type and spark detection command to the spark detection robot. The spark detection robot responds to the detection command from the remote control platform and performs the spark detection task on the exhaust pipes of the vehicle to be inspected within the detection area.

[0018] Compared with existing technologies, the advantages and positive effects of this invention are as follows: The solution provided by this invention offers a robot and system for detecting sparks from the exhaust pipes of transport trucks. This solution utilizes "database-guided precise autonomous positioning" to solve the challenges of vehicle model differentiation and complex spatial positioning, achieving adaptive and high-precision alignment of exhaust outlets from vehicles of different sizes, heights, and orientations. Furthermore, through a unique robotic arm-controlled partial light-shielding method, interference from strong background light such as sunlight is effectively isolated at the physical level, greatly improving the reliability of the detection signal. Simultaneously, by further integrating a sensor self-cleaning module, the optical components are actively kept clean, ensuring the long-term stable operation of the system. Finally, through a mobile platform and fully automated control, the entire process from vehicle identification, navigation, detection execution to result judgment is achieved unmanned operation. This significantly improves detection efficiency and traffic capacity while completely isolating operators from high-temperature and exhaust gas environments, ensuring human-machine safety. It also provides a flexible and efficient solution for multi-detection point, schedulable deployment, systematically overcoming the fundamental deficiencies of existing fixed and handheld detection technologies in terms of anti-interference, adaptability, stability, and automation.

[0019] The present invention designs an integrated detection module and a light-shielding cover driven by an autonomous mobile platform and precisely controlled by a multi-degree-of-freedom robotic arm. The solution employs three steps: prior database positioning → robotic arm precision alignment → light-shielding cover placement. This actively constructs a local dark chamber at the exhaust port, physically isolating the sensor's optical window and the measured area from the external strong light environment. This design significantly avoids direct interference from ambient light at the physical level, enabling the sensor to operate in a stable and pure optical environment. This results in a substantial improvement in the signal-to-noise ratio of the detected signal. Furthermore, combined with the design of a composite optical sensor unit, it effectively improves detection accuracy and virtually eliminates false alarms caused by changes in illumination.

[0020] To address the issue of blind spots in vehicle detection caused by variations in vehicle model, exhaust pipe position, and angle, the solution described in this invention abandons fixed-view scanning or inefficient manual handheld methods. Instead, it innovatively employs a guided positioning strategy based on a priori database. Through a pre-built vehicle-detection point database, the system can automatically plan and control the robot to reach the optimal detection and parking target position based on the identified vehicle model. It also guides the robotic arm to quickly and accurately deliver the detection module to the front of the exhaust port according to a preset recommended posture and offset vector. Simultaneously, combined with the fine-tuning of the tilt angle by the end-effector, the solution described in this invention can adapt to exhaust ports of different heights and orientations, solving the fundamental defects of existing fixed systems, such as fixed viewing angles and poor adaptability.

[0021] To address the problem that existing system sensors are susceptible to performance degradation due to exhaust oil and carbon soot pollution, and lack effective maintenance methods, the solution described in this invention integrates a sensor self-cleaning module into the integrated detection module. This solution can automatically purge and clean the sensor's optical window after each detection task, forming a closed-loop health management mechanism of "detection-cleaning-re-detection". It can actively remove contaminants and maintain the sensor's optimal sensitivity and optical transmittance over a long period of time. This fundamentally solves the pain point of existing technologies where sensor performance deteriorates rapidly over time due to contamination and requires frequent manual maintenance, significantly improving the long-term operational reliability and maintenance-free level of the system. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below: Figure 1 is a schematic diagram of the basic structure of a transport truck exhaust pipe spark detection robot described in the embodiment; Figure 2 is a schematic diagram of the first state (i.e., the lifting platform is raised) of the transport truck exhaust pipe spark detection robot described in the embodiment; Figure 3 is a schematic diagram of the second state (i.e., the lifting platform is raised and the multi-degree-of-freedom robotic arm is deployed) of the transport truck exhaust pipe spark detection robot described in the embodiment; Figure 4 is a schematic diagram of the structure of the transport truck exhaust pipe spark detection robot in the working state described in the embodiment; Figure 5 is an enlarged schematic diagram of partial structure A in Figure 4; Figure 6 is an enlarged schematic diagram of Figure 5. Figure 7 is a schematic diagram of the enlarged partial structure B in the embodiment; Figure 7 is a schematic diagram of the structure of a spark detection system for the exhaust pipe of a transport truck described in the embodiment; wherein, 1, autonomous mobile platform; 1-1, embedded cavity; 1-2, telescopic gate; 2, integrated detection module; 2-1, sensor module; 2-2, light shield; 2-3, sensor self-cleaning module; 2-3-1, micro air tank and micro air pump integrated module; 2-3-2, pipeline; 2-3-3, jet nozzle; 2-4, gimbal; 3, lifting platform; 3-1, telescopic rod; 3-2, electric rotating base; 4, multi-degree-of-freedom robotic arm; 5, detection area; 6, vehicle to be inspected; 6-1, exhaust port; A, first enlarged partial structure; B, second enlarged partial structure. Detailed Implementation

[0023] To better understand the above-mentioned objectives, features and advantages of the present invention, the present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0024] Numerous specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways than those described herein, and therefore the invention is not limited to the specific embodiments disclosed in the following specification.

[0025] Example 1: The following describes in detail, with reference to the accompanying drawings, a spark detection robot for the exhaust pipe of a transport truck as described in Example 1.

[0026] As shown in Figures 1, 2, and 3, a robot for detecting sparks from the exhaust pipes of transport trucks includes an autonomous mobile platform 1, an integrated detection module 2, and a controller. The autonomous mobile platform 1 has an embedded cavity 1-1 on its upper part for storing the integrated detection module 2, and a lifting platform 3 at the bottom of the embedded cavity 1-1. The integrated detection module 2 is connected to the lifting platform 3 via a multi-degree-of-freedom robotic arm 4. The integrated detection module 2 includes a sensor module 2-1 for spark detection and temperature detection, a light-shielding cover 2-2, and a sensor self-cleaning module 2-3. The integrated detection module 2 transmits the collected sensor data to the controller. The controller performs the following processing steps in response to vehicles within the detection area 5. Based on the arrival signal and the identification result of the vehicle 6 to be inspected, and according to the pre-built vehicle-inspection point database, the optimal detection and parking target position of the Mars detection robot, the offset vector of the exhaust port 6-1 relative to the optimal detection and parking target position, and the recommended posture of the multi-degree-of-freedom robotic arm 4 are determined. The optimal detection and parking target position is transmitted to the autonomous mobile platform 1. Based on the navigation and obstacle avoidance strategy built into the autonomous mobile platform 1, the Mars detection robot is controlled to move to the optimal detection and parking target position. Based on the offset vector and the recommended posture of the multi-degree-of-freedom robotic arm 4, the integrated detection module 2 is controlled to reach the preset position of the exhaust port 6-1 to perform data acquisition. Based on the acquired sensor data, the Mars detection result is obtained.

[0027] It should be noted that the autonomous mobile platform 1 can be implemented using a commercially available autonomous mobile robot chassis with target point navigation and obstacle avoidance functions. It integrates at least a Simultaneous Localization and Mapping (SLAM) module, LiDAR, depth camera and other environmental perception sensors and corresponding path planning controllers. It can receive target position instructions sent by the host computer and move autonomously and safely to the designated location. Since such autonomous mobile robot chassis is an existing solution and is widely used, the structure and functional design of the autonomous mobile platform 1 will not be described in detail here.

[0028] In a specific implementation, a telescopic door 1-2 is provided on the top of the embedded cavity 1-1. The telescopic door 1-2 is controlled by the controller. When the Mars detection robot moves to the optimal detection and parking target position, the telescopic door 1-2 is opened, and the multi-degree-of-freedom robotic arm 4 and the integrated detection module 2 are pushed out of the embedded cavity 1-1 through the telescopic rod 3-1 connected between the bottom of the lifting platform 3 and the bottom of the embedded cavity 1-1. After the detection task is completed and the multi-degree-of-freedom robotic arm 4 and the integrated detection module 2 are reset and put back into the embedded cavity 1-1, the telescopic door 1-2 is closed by the controller.

[0029] Understandably, the above design has the following advantages: On the one hand, the integrated detection module 2, as a precision device, is under the strict protection of the embedded cavity 1-1 and the telescopic door 1-2 during the non-working stage, avoiding contamination of the sensor module 2-1 on the integrated detection module 2 due to external environment (such as weather conditions); on the other hand, by adopting an embedded design, the size of the Mars detection robot (especially in height) can be effectively reduced, which can effectively improve the robot's flexibility in moving under the vehicle and greatly reduce the possibility of collision.

[0030] In a specific implementation, as shown in the first partial structure A in Figure 4 and the second partial structure B in Figure 5 (i.e., Figure 6), the integrated detection module 2 is connected to the lifting platform 3 through the multi-degree-of-freedom robotic arm 4. An electric rotating base 3-2 is provided between the multi-degree-of-freedom robotic arm 4 and the lifting platform 3. The electric rotating base 3-2 can realize the horizontal rotation of the multi-degree-of-freedom robotic arm 4, thereby driving the horizontal rotation of the integrated detection module 2.

[0031] In specific implementation, the integrated detection module 2 and the end of the multi-degree-of-freedom robotic arm 4 are equipped with a gimbal 2-4. The gimbal 2-4 can be used to adjust the tilt angle of the integrated detection module 2 in multiple directions to adapt to different tilt angle designs of the exhaust port 6-1.

[0032] In specific implementation, the integrated detection module 2 is provided with a horn-shaped light shield 2-2. The outermost inner diameter of the light shield 2-2 is larger than the outer diameter of the exhaust port 6-1 (the recommended ratio is 1.5:1) to ensure that the light shield 2-2 blocks light when detecting sparks without affecting the normal exhaust of the exhaust port 6-1.

[0033] In one or more embodiments, the multi-degree-of-freedom robotic arm 4 adopts a two-stage robotic arm structure. The sensor self-cleaning module 2-3 of the integrated detection module 2 includes a micro air tank and a micro air pump integrated module 2-3-1. The micro air tank and the micro air pump integrated module 2-3-1 are connected to the air nozzle 2-3-3 located at the position of the sensor module 2-1 through a pipeline 2-3-2. The micro air tank and the micro air pump integrated module 2-3-1 are fixed on the second stage of the multi-degree-of-freedom robotic arm 4.

[0034] In a specific implementation, the sensor module 2-1 includes a composite optical sensor unit and a temperature sensor unit, wherein the composite optical sensor unit includes a first sensor sensitive to the characteristic spectrum of Martian light and a second sensor responding to ambient white light.

[0035] It should be noted that the first sensor is a photoelectric sensor, such as a high-speed narrowband near-infrared photodiode, whose core component is an indium gallium arsenide photodiode. Its operating wavelength is 0.9 to 1.7 μm (short-wave infrared). The reason for choosing this design is that high-temperature particles have high radiation intensity in this wavelength band and are less affected by environmental interference. In addition, the sensor's response time can reach the nanosecond level, which can capture instantaneous sparks. At the same time, it can be used with a narrowband filter to further shield non-target radiation.

[0036] In one or more embodiments, the first sensor may also employ existing sensor devices such as ultraviolet flame detectors and photomultiplier tubes.

[0037] In one or more embodiments, the second sensor may be a broadband visible light sensor, such as a silicon photodiode with a spectral response range of 400 to 700 nm, which can effectively cover visible light.

[0038] In practice, the first and second sensors are installed side by side in the same optical window to ensure that the two sensors monitor the same spatial area.

[0039] In specific implementation, the process of obtaining the Mars detection result based on the collected sensor data involves the following steps: acquiring real-time data from the first and second sensors. and (1) Baseline correction and peak extraction: Based on the obtained sensor data, the steady-state background values ​​corresponding to the two sensors are calculated respectively, as follows: in, This represents the steady-state background value corresponding to the first sensor. This represents the steady-state background value corresponding to the second sensor. The moment when the data detected by the first sensor exceeds a preset threshold.

[0040] Based on the obtained steady-state background value, the signal peak value is extracted (from...). Starting with M consecutive points, specifically represented as follows: in, The peak value of the signal from the first sensor. The peak value of the signal from the second sensor.

[0041] (2) Based on the obtained signal peak value, calculate the spectral characteristic ratio, specifically expressed as follows: Spectral characteristic ratio: in, For a very small number (the scheme described in this embodiment adopts...) (), to prevent the denominator from being 0.

[0042] (3) Based on the obtained signal peak value of the first sensor, combined with the spectral feature ratio, determine whether it is a spark; specifically, when the signal peak value of the first sensor is less than the first preset threshold (e.g., 100 millivolts), it is determined to be noise; when it is greater than the first preset threshold, determine whether the spectral feature ratio is greater than the second preset threshold. If it is, it is determined to be a spark. If it is greater than the third threshold but less than or equal to the second threshold, it is determined to be a possible spark and the next round of detection is performed. If it is less than or equal to the third threshold, it is determined not to be a spark.

[0043] In practice, after the current task is completed (i.e., the controller determines whether there are sparks), the controller controls the sensor self-cleaning module 2-3 to open the electronic valve located at the outlet of the miniature gas tank, and the high-speed gas released by the jet nozzle 2-3-3 cleans the sensor module 2-1.

[0044] In one or more embodiments, a pressure sensor is installed at the air inlet of the sensor self-cleaning module 2-3 to detect the air pressure of the micro air tank in real time. When the air pressure is lower than a preset threshold, the controller controls the micro air pump integrated module 2-3-1 to pressurize the micro air tank.

[0045] In specific implementation, based on a pre-built vehicle-inspection point database, the optimal detection and parking target position of the Mars detection robot, the offset vector of the exhaust port 6-1 relative to the optimal detection and parking target position, the unit vector of the orientation of the exhaust port 6-1, and the recommended posture of the multi-degree-of-freedom robotic arm 4 are determined. The optimal detection and parking target position is transmitted to the autonomous mobile platform 1. Based on the navigation and obstacle avoidance strategies built into the autonomous mobile platform 1, the Mars detection robot is controlled to move to the optimal detection and parking target position. Based on the offset vector and the recommended posture of the multi-degree-of-freedom robotic arm 4, the integrated detection module is controlled to... 2. Data acquisition is performed at the preset position of exhaust port 6-1; based on the acquired sensor data, the spark detection results are obtained, specifically including the following processing steps: (1) Vehicle-detection point database construction. The scheme described in this embodiment takes into account that the transport vehicles in the factory area usually use several (a small number) fixed types of vehicles for product transportation and the efficiency of spark detection. It abandons the traditional method of robot autonomous inspection that relies entirely on machine vision and artificial intelligence algorithms. Instead, it constructs a vehicle-detection point database and stores the ideal detection parameter data of each vehicle model in advance. Specifically, it includes: the precise three-dimensional coordinates of exhaust port 6-1 of each vehicle model relative to the rear wheel axle center (reference point) ( The optimal target position for robot detection and parking (e.g., 1.5m behind the vehicle, directly opposite the center line of exhaust port 6-1), the offset vector of exhaust port 6-1 relative to the optimal target position, the unit vector of exhaust port 6-1 orientation, the recommended posture of multi-degree-of-freedom robotic arm 4, and the safe distance between the light shield 2-2 and exhaust port 6-1 (the purpose is to ensure the light shielding effect while ensuring normal exhaust of exhaust port 6-1).

[0046] (2) Ideal detection parameter data acquisition is based on the pre-built vehicle-detection point database. Combined with the type of vehicle 6 to be inspected in detection area 5, the three-dimensional coordinates of the exhaust port 6-1 of the vehicle 6 to be inspected relative to the center of the rear wheel axle, the optimal detection parking target position of the robot, the offset vector of the exhaust port 6-1 relative to the optimal detection parking target position, the unit vector of the direction of the exhaust port 6-1, the recommended posture of the multi-degree-of-freedom robotic arm 4, and the safe distance between the light shield 2-2 and the exhaust port 6-1 are obtained. (3) The robot moves to the target position and transmits the optimal detection parking target position to the autonomous mobile platform 1. Based on the navigation and obstacle avoidance strategy built into the autonomous mobile platform 1, the Mars detection robot is controlled to move to the optimal detection parking target position. It can be understood that the existing autonomous mobile platform 1, based on its hardware settings and built-in algorithms, can accurately move to the preset target position even if there is a certain deviation in the parking position of different drivers in detection area 5. Therefore, the target position movement process of the robot will not be described in detail here.

[0047] (4) The robotic arm is precisely positioned and the light shield 2-2 is in place to obtain the optimal target position for the robot to detect and park the current vehicle model. This is combined with the offset vector of the exhaust port 6-1 relative to the optimal target position and the unit vector of the direction in which the exhaust port 6-1 is oriented. Recommended posture of multi-degree-of-freedom robotic arm 4 and safe distance between the light shield 2-2 and the exhaust port 6-1. Using the robot's parking position as a reference, the precise coordinates of exhaust port 6-1 in the world coordinate system are calculated by adding the aforementioned offset vector. The multi-degree-of-freedom robotic arm 4 is controlled to move to the recommended posture, achieving coarse alignment between the integrated detection module 2 and the exhaust port 6-1. Based on the calculated precise coordinates of the exhaust port 6-1 in the world coordinate system, combined with the safe distance between the light shield 2-2 and the exhaust port 6-1 and the unit vector of the exhaust port 6-1's orientation, the target position of the light shield 2-2 at the end of the robotic arm is calculated. Specifically, it is expressed as follows: Control the end effector of the robotic arm (i.e., the mounting point of the light shield 2-2) to move to the target position. And through the gimbal 2-4, the central axis of the sunshade 2-2 is aligned with the unit vector of the exhaust port 6-1. Overlap as much as possible.

[0048] (5) Perform sensor data acquisition, and process the data based on the aforementioned Martian detection process to obtain Martian detection results. Issue an alarm based on the detection results, such as a voice broadcast saying "Martians are present, do not enter".

[0049] (6) After completing the current detection, perform self-cleaning of sensor module 2-1.

[0050] In one or more embodiments, considering that ammonia may explode under high temperature conditions, the temperature of the exhaust port 6-1 is detected by the temperature sensor in the sensor module 2-1 of the integrated detection module 2. When the temperature exceeds the preset threshold, an alarm is issued, for example, in the form of a voice broadcast: "Temperature too high, do not enter".

[0051] Example 2: In one or more embodiments, as shown in Figure 7, this embodiment provides a spark detection system for the exhaust pipes of a transport truck, including the aforementioned spark detection robot for the exhaust pipes of a transport truck, a monitoring module, and a remote control platform; wherein: the monitoring module is used to collect images of the detection area 5 in real time and transmit them to the remote control platform; the remote control platform is used to receive images of the detection area 5 and determine the arrival status and vehicle type of the vehicle to be inspected 6 based on a built-in image recognition algorithm; when the vehicle to be inspected 6 arrives, the platform sends the vehicle type and spark detection command to the spark detection robot; the spark detection robot responds to the detection command from the remote control platform and executes the spark detection task of the exhaust pipes of the vehicle to be inspected 6 in the detection area 5.

[0052] In specific implementation, the image recognition algorithm can be a deep learning-based image recognition algorithm, such as CNN, RNN and VGG16 models. Essentially, it is to obtain the recognition of whether the vehicle has entered the detection area 5 and the vehicle type. Since the traditional model used in this embodiment is the input of the captured image and the output is the recognition of whether the vehicle has entered the detection area 5 and the vehicle type, it can be achieved by those skilled in the art with the help of existing models. Therefore, the image recognition algorithm will not be described in detail.

[0053] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications or equivalent changes made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A robot for detecting sparks from the exhaust pipe of a transport truck, characterized in that, The system includes an autonomous mobile platform, an integrated detection module, and a controller. The autonomous mobile platform has an embedded cavity at its upper part for storing the integrated detection module, and a lifting platform at its bottom. The integrated detection module is connected to the lifting platform via a multi-degree-of-freedom robotic arm. The integrated detection module includes a sensor module for detecting sparks and temperature, a light-shielding cover, and a sensor self-cleaning module. The integrated detection module transmits the collected sensor data to the controller. The controller performs the following processing steps: in response to a vehicle arrival signal within the detection area and the vehicle identification result, it determines the vehicle based on a pre-built vehicle... A vehicle-detection point database is used to determine the optimal detection and parking target position for the Mars robot, the offset vector of the exhaust port relative to the optimal detection and parking target position, and the recommended posture of the multi-degree-of-freedom robotic arm. The optimal detection and parking target position is transmitted to the autonomous mobile platform. Based on the navigation and obstacle avoidance strategies built into the autonomous mobile platform, the Mars robot is controlled to move to the optimal detection and parking target position. Based on the offset vector and the recommended posture of the multi-degree-of-freedom robotic arm, the integrated detection module is controlled to reach the preset position of the exhaust port to perform data acquisition. Based on the acquired sensor data, the Mars detection result is obtained.

2. The spark detection robot for the exhaust pipe of a transport truck as described in claim 1, characterized in that, The top of the embedded cavity is equipped with a telescopic door, which is controlled by the controller. When the Mars detection robot moves to the optimal detection and parking target position, it opens and extends, and the multi-degree-of-freedom robotic arm and integrated detection module are pushed out of the embedded cavity through the telescopic rod connected between the bottom of the lifting platform and the bottom of the embedded cavity. Furthermore, after completing the inspection task, the multi-degree-of-freedom robotic arm and the integrated inspection module reset and enter the embedded cavity, and then the telescopic gate is closed by the controller.

3. The spark detection robot for the exhaust pipe of a transport truck as described in claim 1, characterized in that, The integrated detection module is equipped with a horn-shaped light shield, the outermost inner diameter of which is larger than the outer diameter of the exhaust port.

4. The spark detection robot for the exhaust pipe of a transport truck as described in claim 1, characterized in that, The multi-degree-of-freedom robotic arm adopts a two-stage robotic arm structure. The sensor self-cleaning module of the integrated detection module includes a micro air tank and a micro air pump integrated module. The micro air tank and the micro air pump integrated module are connected to the air nozzle located at the sensor module position through a pipeline.

5. The spark detection robot for the exhaust pipe of a transport truck as described in claim 1, characterized in that, The integrated detection module is connected to the lifting platform via a multi-degree-of-freedom robotic arm, wherein an electric rotating base is provided between the multi-degree-of-freedom robotic arm and the lifting platform.

6. The spark detection robot for the exhaust pipe of a transport truck as described in claim 1, characterized in that, The integrated detection module and the end of the multi-degree-of-freedom robotic arm are equipped with a gimbal. The gimbal can be used to adjust the tilt angle of the integrated detection module in multiple directions to adapt to different tilt angle designs of the exhaust port.

7. The spark detection robot for the exhaust pipe of a transport truck as described in claim 1, characterized in that, The sensor module includes a composite optical sensor unit and a temperature sensor unit, wherein the composite optical sensor unit includes a first sensor sensitive to the characteristic spectrum of Martian light and a second sensor responding to ambient white light.

8. The spark detection robot for the exhaust pipe of a transport truck as described in claim 7, characterized in that, The process of obtaining the Mars detection result based on the collected sensor data involves the following steps: acquiring real-time data from the first and second sensors; calculating the steady-state background values ​​corresponding to the two sensors based on the acquired sensor data; extracting the signal peak value based on the acquired steady-state background value; and calculating the spectral feature ratio based on the acquired signal peak value. Based on the obtained signal peak value of the first sensor, combined with the spectral feature ratio, it is determined whether it is a Mars particle. Specifically, when the signal peak value of the first sensor is less than a first preset threshold, it is determined to be noise. When it is greater than the first preset threshold, it is determined whether the spectral feature ratio is greater than a second preset threshold. If it is, it is determined to be a Mars particle. If it is greater than a third threshold but less than or equal to the second threshold, it is determined to be a possible Mars particle and the next round of detection is performed. If it is less than or equal to the third threshold, it is determined not to be a Mars particle.

9. The spark detection robot for the exhaust pipe of a transport truck as described in claim 1, characterized in that, The integrated detection module performs data acquisition when it reaches the preset position of the exhaust port. Specifically, it obtains the optimal detection and parking target position of the robot corresponding to the current vehicle model, and combines the offset vector of the exhaust port relative to the optimal detection and parking target position, the unit vector of the exhaust port orientation, the recommended posture of the multi-degree-of-freedom robotic arm, and the safe distance between the sunshade and the exhaust port. Using the robot's parking position as a reference, the precise coordinates of the exhaust port in the world coordinate system are calculated by adding the aforementioned offset vector. The multi-degree-of-freedom robotic arm is controlled to move to the recommended posture to achieve coarse alignment between the integrated detection module and the exhaust port. Based on the calculated precise coordinates of the exhaust port in the world coordinate system, combined with the safe distance between the light shield and the exhaust port and the unit vector of the exhaust port's orientation, the target position of the light shield at the end of the robotic arm is calculated. The end of the robotic arm is controlled to move to the target position, and the central axis of the light shield is aligned with the unit vector of the exhaust port's orientation via gimbal control. Sensor data acquisition is performed, and data processing is conducted based on the aforementioned spark detection process to obtain the spark detection result.

10. A spark detection system for the exhaust pipe of a transport truck, characterized in that, The invention includes a transport truck exhaust pipe spark detection robot, a monitoring module, and a remote control platform as described in any one of claims 1-9; wherein: the monitoring module is used to collect images of the detection area in real time and transmit them to the remote control platform; the remote control platform is used to receive images of the detection area and determine the arrival status and vehicle type of the vehicle to be inspected based on a built-in image recognition algorithm; when the vehicle to be inspected arrives, the platform sends the vehicle type and spark detection command to the spark detection robot; the spark detection robot responds to the detection command from the remote control platform and performs the task of detecting sparks from the exhaust pipe of the vehicle to be inspected in the detection area.

Images