An intelligent perception system and method for a loader

Abstract

The application discloses an intelligent sensing system and method for a loader, which is composed of an intelligent sensing system control unit and two front and rear subsystems. Each subsystem comprises a base, a multi-degree-of-freedom mechanical arm, a camera and a laser radar sensing device at the end, wherein the base is used for fixing the intelligent sensing system on the loader vehicle body; the multi-degree-of-freedom mechanical arm can make the sensing device at the end reach any position within the maximum radius range of the mechanical arm through swinging and rotating and keep stable; the camera and the laser radar perform the environment sensing function. Compared with the traditional scheme of fixing the camera and the radar on the vehicle, the intelligent sensing system can automatically evaluate the sufficiency of the current sensing information, if it is judged that there may be a sensing blind area next time, the mechanical arm will be started to move in advance to reduce the existence of the sensing blind area as much as possible, so as to avoid the collision risk under the premise of not increasing the number of sensing devices too much, and provide protection for the driving and operation safety of the loader.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of engineering machinery, and particularly relates to an intelligent sensing system and method for a loader. BACKGROUND

[0002] Loaders are widely used and are responsible for operations such as loading, unloading and stacking materials at places such as ports, mixing stations, coal mines, steel plants, power plants and coking plants. Due to the harsh and complex working environment, heavy and tiring work tasks, and sudden and variable work rhythm, the loader driver has a heavy workload and is prone to fatigue, which makes it difficult to respond to changes in the environment or events around the loader in a timely manner, resulting in safety accidents.

[0003] Although intelligent auxiliary loaders, remote control loaders and unmanned loaders exist in the market to improve the automation and intelligence level of equipment, none of them have the sensing device fixed on the vehicle body. This approach often has the following drawbacks: 1) due to the insufficient number of sensing devices, combined with the large size of the loader and the large movement range of the working device, it is easy to cause a sensing blind area, thereby affecting the safety of the operation or the accuracy of the loading and unloading; 2) increasing the sensing devices can alleviate the aforementioned problems to a certain extent, but it increases the cost of the whole machine and causes certain difficulties in mechanical arrangement, installation and equipment maintenance. SUMMARY

[0004] The present application provides an intelligent sensing system and method for a loader, which is composed of an intelligent sensing system control unit and two front and rear subsystems. Each subsystem includes a base, a multi-degree-of-freedom robot arm, a camera at the end and a laser radar sensing device. The base is used to fix the intelligent sensing system on the loader vehicle body; the multi-degree-of-freedom robot arm can move the sensing device at the end to any position within the maximum radius range of the robot arm and keep it stable through swinging and rotating; the camera and laser radar perform environmental sensing functions. Compared with the traditional scheme of fixing the camera and radar on the vehicle, the intelligent sensing system can automatically evaluate the sufficiency of the current sensing information. If it is judged that there may be a sensing blind area next time, the robot arm will be moved in advance to minimize the existence of the sensing blind area, thereby avoiding the risk of collision without increasing the number of sensing devices, providing protection for the safety of the loader driving and operation. Through the system and method of the present application, the loader can control the sensing device to flexibly detect the surrounding environment and operation situation from different poses and angles, thereby achieving the purpose of eliminating the sensing blind area based on a small number of sensing devices.

[0005] Technical solution: An intelligent sensing system for a loader, comprising: an intelligent sensing system control unit, and two front and rear subsystems 1 and 2.

[0006] The intelligent perception system control unit is responsible for the fusion processing of the surrounding environment information collected by the two subsystems, and according to the real-time decision evaluation of the loader control decision system, the next possible perception blind area encountered by the loader is determined, and then the motion control of the subsystems 1 and 2 is performed;

[0007] The subsystem 1 is located at the front of the top axis of the loader cab, and the subsystem 2 is located at the rear of the top axis of the loader cab, both of which include a base, a multi-degree-of-freedom mechanical arm, and a camera and a laser radar perception device at the end of the mechanical arm, and the multi-degree-of-freedom mechanical arm is installed on the loader through the base;

[0008] The subsystem 1 is responsible for the environment perception of the front part of the vehicle body in the forward and lateral directions, and the subsystem 2 is responsible for the environment perception of the rear part of the vehicle body in the rearward and lateral directions.

[0009] Further, the intelligent perception system control unit includes a communication module and an operation processing module;

[0010] The communication module is responsible for: 1) receiving the surrounding environment information collected by the camera and the laser radar of the two subsystems through the network cable; 2) receiving the real-time decision information from the loader control decision system through the network cable or other high-speed communication mode; 3) issuing corresponding motion control instructions to the two subsystems through the CAN bus;

[0011] The operation processing module is responsible for: 1) synchronizing, denoising, comparing and other perception fusion processing of the surrounding environment information collected by the camera and the laser radar of the two subsystems, so that the loader has a clear perception of the surrounding environment at the current time; 2) determining the next action intention of the loader according to the real-time decision information of the loader control decision system, and evaluating whether the loader will encounter a certain perception blind area at the next time when it is in a new pose, in combination with the current perceived environment information; 3) if the evaluation result is that the perception blind area may be encountered, the motion parameters and attitude of the mechanical arm need to be adjusted in real time to eliminate the blind area of the camera and the laser radar as much as possible; 4) calculating a damping control model according to the vibration of the vehicle in real time.

[0012] Further, the lower part of the base in the subsystem is responsible for fixing the subsystem on the loader, and the motor in the base is responsible for driving the upper platform to rotate 360°, so that the multi-degree-of-freedom mechanical arm on the base can rotate 360°;

[0013] Each joint of the multi-degree-of-freedom mechanical arm can swing within a certain range of arc around the joint, so that the camera and the laser radar perception device at the end of the mechanical arm can reach any position within the maximum radius range of the mechanical arm;

[0014] The laser radar perception device includes a semi-solid laser radar and a mechanical multi-line laser radar;

[0015] The end of the last section of the multi-degree-of-freedom robot arm can rotate 360°, ensuring that the three types of sensing devices, i.e., a high-definition camera, a semi-solid laser radar, and a mechanical multi-line laser radar, have a larger and more complete angle coverage.

[0016] The multi-degree-of-freedom robot arm is provided with active damping stabilizing devices at the connection with the base, the joints, and the end, and together with all the motors of the robot arm, maintains the stability of the sensing device end under the algorithm instructions of the intelligent sensing system control unit.

[0017] Further, the communication module includes a USB-to-CAN device and a network card.

[0018] The USB-to-CAN device is connected to the intelligent sensing system control unit through a USB interface and is responsible for converting the control instructions of the control unit to the subsystem into CAN signals.

[0019] The network card is connected with the camera and the laser radar, so that the intelligent sensing system control unit can obtain the sensing data in real time.

[0020] Further, the base includes a CAN communication unit for communication with the intelligent sensing system control unit, a motor control unit, and a calibration parameter storage unit.

[0021] The CAN communication unit is responsible for receiving instructions from the intelligent sensing system control unit and feeding back the motion state of the robot arm.

[0022] The motor control unit is responsible for motion control of all the motors of the robot arm.

[0023] The calibration parameter storage unit is responsible for storing the motion parameters of each motor in different poses.

[0024] An intelligent sensing method for a loader includes:

[0025] When the intelligent sensing control unit combines the sensing information of the surrounding environment at the current time t0, the current pose of the loader, and the motion intention information of the loader, it can predict and evaluate the new pose lp1 of the loader at the next time t1.

[0026] Based on the new pose lp1 of the loader, combined with the poses s0 of the current sensing devices of the intelligent sensing system, the latest poses s1 of the sensing devices at the next time can be obtained.

[0027] Based on the new poses s1 of the sensing devices, combined with the sensing methods and fields of view of the sensing devices and the sensing information at the current time, the sensing information at the next time can be comprehensively predicted and evaluated to determine whether there is a sensing blind area in some areas.

[0028] If the existence of the sensing blind area is evaluated, the mechanical arm is immediately driven to move the sensing device to several specified areas for environment observation and scanning, so as to eliminate or reduce the sensing blind area at t1 as much as possible.

[0029] In further embodiments, the application scenarios of the intelligent sensing method for the loader include:

[0030] When the loader is in normal driving, compared with the conventional method of fixing the sensing device, the intelligent sensing system can drive the mechanical arm to perform dead angle-free observation on the vicinity of the bucket and the vicinity of the boom, so as to eliminate the near view blind area.

[0031] When the loader is paying attention to the front and rear environment, the mechanical arm can be driven to periodically perform blind area supplementing scanning observation on the areas on the left and right sides of the vehicle body, and if a person or other object is found to suddenly intrude into the safety range on both sides of the loader, the obstacle avoidance and anti-collision measures are immediately taken.

[0032] When there is a barrier such as a fence near the loader and the barrier has a limited height, the mechanical arm can be stretched upward to make the sensing device have a higher and better view, so that the situation behind the barrier can be understood early, and safety precautions can be made in advance.

[0033] When the loader completes the bucketing action and plans to collect the bucket, compared with the conventional method of fixing the sensing device, the intelligent sensing system can make a more accurate judgment on the bucketing fullness rate by adjusting the observation distance and the viewing angle of the sensing device.

[0034] Since the lifting of the boom during unloading of the loader can easily block the forward view of the sensing device, the mechanical arm can be driven to detect and sense the unloading place from other angles, so as to judge the material stacking condition in the full bucket and accordingly select the best unloading point, thereby preventing uneven material accumulation or avoiding the problem of material spilling due to inaccurate unloading position.

[0035] When the loader completes the unloading plan and collects the bucket, the conventional method of fixing the sensing device is difficult to detect whether there is residual material in the bucket of the loader, and the intelligent sensing system can control the mechanical arm to make the sensing device accurately observe the material condition in the bucket.

[0036] When the unmanned loader is working in a tunnel or other space with a relatively small space, the intelligent sensing system can control and adjust the pose of the mechanical arm in real time according to the actual movable space range, so as to ensure that the mechanical arm meets the detection requirements to the maximum extent under the premise of safety and no collision.

[0037] Advantages: Compared with the prior art, the present application has the following advantages:

[0038] The perception device is no longer fixed at a certain position of the loader body, but can move flexibly according to actual needs, and can perceive and observe the environment in a larger range in a richer, more forward-looking and more scientific and reasonable pose.

[0039] The perception blind area is eliminated with as few perception devices as possible, and the problems of rising overall cost of the machine, difficult mechanical arrangement and installation and device maintenance caused by too many perception devices are avoided. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 The figure shows a schematic diagram of the intelligent perception subsystem of the application;

[0041] Figure 2 The figure shows a schematic diagram of the perception range of the intelligent perception subsystem 1 and 2 of the application;

[0042] Figure 3 The figure shows a schematic diagram of the intelligent perception system scanning and observing the left and right sides of the vehicle body;

[0043] Figure 4 The figure shows a schematic diagram of the intelligent perception method when facing obstacles;

[0044] Figure 5 The figure shows a schematic diagram of the intelligent perception method when facing unloading;

[0045] Figure 6 The figure shows a schematic diagram of the intelligent perception system of the loader of the application. EMBODIMENTS

[0046] In order to more fully understand the technical content of the application, the technical solutions of the application will be further introduced and described below in combination with specific embodiments, but are not limited thereto.

[0047] As shown in the figure, an intelligent perception system for a loader comprises an intelligent perception system control unit, and two subsystems 1 and 2 in front and back; Figure 6 The intelligent perception system control unit is responsible for fusion processing of the surrounding environment information collected by the two subsystems, and according to the real-time decision evaluation of the loader control decision system, the possible perception blind area encountered by the loader next is evaluated, and then the motion control of the subsystems 1 and 2 is performed;

[0048] The subsystem 1 is located at the front of the axle center of the top of the loader cab, and the subsystem 2 is located at the rear of the axle center of the hood of the rear vehicle body of the loader, both of which comprise a base, a multi-degree-of-freedom mechanical arm, and a camera and a laser radar perception device at the end of the mechanical arm, and the multi-degree-of-freedom mechanical arm is installed on the loader through the base;

[0049]

[0050] ​Subsystem 1 is responsible for the front body forward, lateral range of environmental perception, and subsystem 2 is responsible for the rear body rear, lateral range of environmental perception.

[0051] In further embodiments, the intelligent perception system control unit comprises a communication module and an operation processing module;

[0052] The communication module is responsible for: 1) receiving the surrounding environment information collected by the cameras and laser radars of the two subsystems through the network cable; 2) receiving real-time decision information from the loader control decision system through the network cable or other high-speed communication methods; 3) issuing corresponding motion control instructions to the two subsystems through the CAN bus;

[0053] The operation processing module is responsible for: 1) synchronizing, denoising, comparing, and other perception fusion processing of the surrounding environment information collected by the cameras and laser radars of the two subsystems, so that the loader has a clear understanding of the current surrounding environment; 2) determining the next action intention of the loader according to the real-time decision information of the loader control decision system, and evaluating whether the loader will encounter a certain perception blind area at the new pose in the next moment in combination with the current perceived environment information; 3) if the evaluation result is that it may encounter a perception blind area, real-time calculation is needed to adjust the motion parameters and posture of the mechanical arm to eliminate the blind area of the camera and laser radar as much as possible; 4) real-time calculation of a damping control model according to the vehicle vibration condition.

[0054] As shown in Figure 1 The intelligent perception subsystem base, multi-degree-of-freedom mechanical arm, and camera and laser radar perception devices at the end of the mechanical arm are installed on the loader through the base; the multi-degree-of-freedom mechanical arm on the base can realize 360° rotation, and each joint of the multi-degree-of-freedom mechanical arm can swing within a certain range of arc around the joint, so that the camera and laser radar perception devices at the end of the mechanical arm can reach any position within the maximum radius range of the mechanical arm; the end of the last section of the multi-degree-of-freedom mechanical arm can rotate 360°, which ensures that the camera and laser radar perception devices can reach any position within the maximum radius range of the mechanical arm; the laser radar perception devices include semi-solid laser radars and mechanical multi-line laser radars, which ensure that the three types of perception devices, i.e., high-definition cameras, semi-solid laser radars, and mechanical multi-line laser radars, have a larger and more complete angle coverage; the multi-degree-of-freedom mechanical arm is connected to the base, joints, and end, and is equipped with active damping stabilizing devices, which, together with all the motors of the mechanical arm, maintain the stability of the perception device end under the algorithm instructions of the intelligent perception system control unit.

[0055] In further embodiments, the communication module includes a USB-to-CAN device and a network card;

[0056] The USB-CAN device is connected with the intelligent perception system control unit through the USB interface, and is responsible for converting the control instructions of the control unit to the subsystem into CAN signals;

[0057] The network card is connected with the camera and the laser radar, so that the intelligent perception system control unit can obtain the sensing data in real time;

[0058] In a further embodiment, the base includes a CAN communication unit, an electric motor control unit, and a calibration parameter storage unit, which communicate with the intelligent perception system control unit;

[0059] The CAN communication unit is responsible for receiving the instructions of the intelligent perception system control unit and feeding back the motion state of the robot arm to the intelligent perception system control unit;

[0060] The electric motor control unit is responsible for motion control of all electric motors of the robot arm;

[0061] The calibration parameter storage unit is responsible for storing the motion parameters of each electric motor in different poses;

[0062] As shown in Figure 2 The intelligent perception systems 1 and 2 are responsible for sensing the front and rear areas of the loader and the left and right sides of the respective areas, including the vicinity of the loader body, and truly achieving full-area and dead-angle-free coverage.

[0063] As shown in Figure 3 The intelligent perception system periodically performs blind area scanning and observation on the areas on the left and right sides of the vehicle body, and is always alert to whether a person or other object suddenly intrudes into the safety range on both sides of the vehicle body, so as to take obstacle avoidance and collision avoidance measures in advance and ensure driving safety.

[0064] As shown in Figure 4 The loader can eliminate the sensing blind area through the intelligent perception system, sense the environment behind the surrounding obstacles before turning, and make sufficient preparations for safe driving.

[0065] As shown in Figure 5 The left side is a side view of the loader during unloading, and the right side is a top view of the material hopper during unloading. Different color blocks in the top view represent materials with different stacking heights, and smaller color blocks represent higher stacking heights. From the side view, it can be seen that since the position of the material hopper is high, the intelligent perception device of the loader can obtain the observation field of the unloading hopper by stretching forward and upward, and determine that the loader should preferentially select to unload the material to the right of the material hopper during this unloading.

[0066] An intelligent perception method for a loader, comprising:

[0067] When the intelligent perception control unit combines the perception information of the surrounding environment at the current time t0, the current pose of the loader, and the motion intention information of the loader, it can predict and evaluate the new pose lp1 of the loader at the next time t1;

[0068] Based on the new pose lp1 of the loader, combined with the current pose s0 of each perception device of the intelligent perception system, the latest pose s1 of each perception device at the next time can be obtained;

[0069] Based on the new pose s1 of each perception device, combined with the perception method and field of view of each perception device and the perception information at the current time, the perception information at the next time can be comprehensively predicted and evaluated to determine whether there is a perception blind area in some area;

[0070] If it is determined that there is a perception blind area, the mechanical arm is immediately driven to move each perception device to a number of designated areas for environmental observation and scanning, so as to eliminate or reduce the perception blind area at time t1 as much as possible.

[0071] In further embodiments, the application scenarios of the intelligent perception method for the loader include:

[0072] When the loader is driving normally, compared with the conventional method of fixed perception devices, the intelligent perception system can drive the mechanical arm to perform dead-angle-free observation on the vicinity of the bucket and the vicinity of the boom, so as to eliminate the near-field blind area;

[0073] When the loader is paying attention to the front and rear environmental objects, the mechanical arm can be driven to periodically perform blind area scanning observation on the areas on both sides of the vehicle body. If a person or other object is found to suddenly intrude into the safety range on both sides of the loader, an obstacle avoidance and collision avoidance measure is immediately taken;

[0074] When there is an obstacle such as a fence near the loader and the height of the obstacle is limited, the mechanical arm can be stretched upward to make the perception device have a higher and better field of view, so that the situation behind the obstacle can be understood early and safety precautions can be taken in advance;

[0075] When the loader completes the bucketing action plan and starts to collect the bucket, compared with the conventional method of fixed perception devices, the intelligent perception system can make a more accurate judgment on the bucketing fullness rate by adjusting the observation distance and angle of the perception device;

[0076] Since the lifting of the boom during unloading of the loader can easily block the forward field of view of the perception device, the mechanical arm can be driven to detect and perceive the unloading place from other angles, so as to judge the material stacking condition in the full bucket and accordingly select the best unloading point, thereby preventing uneven material accumulation or avoiding the problem of material spilling due to inaccurate unloading position;

[0077] When the loader completes the unloading plan, the traditional way of fixing the sensing device is difficult to detect whether there is residual material in the loader bucket, and the intelligent sensing system can control the mechanical arm to accurately observe the material in the bucket.

[0078] When the unmanned loader is working in a relatively narrow space such as a tunnel, the intelligent sensing system can control and adjust the pose of the mechanical arm in real time according to the actual movable space range, so as to ensure that the mechanical arm meets the detection requirements to the maximum extent under the premise of safety and collision-free.

[0079] The above only describes the preferred embodiments of the present application and is not intended to limit the present application. Any modification, equivalent replacement or improvement made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. An intelligent sensing system for a loader, characterized in that: It includes an intelligent sensing system control unit, as well as two subsystems, 1 and 2. The intelligent perception system control unit is responsible for fusing the surrounding environment information collected by the two subsystems, and assessing the perception blind spots that the loader may encounter next based on the real-time decision of the loader control decision system, and then performing motion control on subsystems 1 and 2. Subsystem 1 is located at the leading edge of the axis on the top of the loader cab, and subsystem 2 is located at the rear edge of the axis of the loader's rear body cover. Both of them include a base, a multi-degree-of-freedom robotic arm, and a camera and lidar sensing device at the end of the robotic arm. Subsystem 1 is responsible for environmental perception in the front and lateral ranges of the front of the vehicle body, and subsystem 2 is responsible for environmental perception in the rear and lateral ranges of the rear of the vehicle body. The intelligent sensing system control unit includes a communication module and a computing module; The communication module is responsible for: 1) receiving ambient environmental information collected by cameras and lidar from the two subsystems via network cable; 2) receiving real-time decision information from the loader control decision system via network cable or other high-speed communication methods; and 3) sending corresponding motion control commands to the two subsystems via CAN bus. The computation and processing module is responsible for: 1) performing synchronization, noise reduction, comparison and other perception fusion processing on the surrounding environment information collected by the cameras and lidar of the two subsystems, so that the loader can have a clear perception and understanding of the surrounding environment at the current moment; 2) Determine the loader's next action intention based on the real-time decision information of the loader control decision system, and assess whether the loader will encounter a certain perception blind spot when it is in a new position at the next moment, in combination with the currently perceived environmental information; 3) If the assessment result is that a perception blind spot may be encountered, calculate in real time how to adjust the robotic arm motion parameters and posture to eliminate the blind spots of the camera and lidar as much as possible; 4) Calculate the vibration reduction control model in real time based on the vehicle vibration.

2. The intelligent sensing system for a loader according to claim 1, characterized in that... ； The lower part of the base in the subsystem is responsible for fixing the subsystem to the loader, and the motor inside the base is responsible for driving the upper platform to rotate 360°, thereby enabling the multi-degree-of-freedom robotic arm on the base to rotate 360°. In a multi-degree-of-freedom robotic arm, each joint can swing within a certain arc range, allowing the camera and lidar sensing devices at the end of the robotic arm to reach any position within the maximum radius of the robotic arm. LiDAR sensing devices include semi-solid-state LiDAR and mechanical multi-line LiDAR; The end effector of the last section of the multi-degree-of-freedom robotic arm can rotate 360°, ensuring that the three types of sensing devices at the end effector—high-definition camera, semi-solid-state LiDAR, and mechanical multi-line LiDAR—have a larger and more comprehensive angular coverage. The multi-degree-of-freedom robotic arm is equipped with active vibration damping and stabilization devices at the connection points with the base, joints, and end effector. Together with all the motors of the robotic arm, these devices maintain the stability of the sensing device under the algorithm instructions of the intelligent sensing system control unit.

3. The intelligent sensing system for a loader according to claim 1, characterized in that: The communication module includes a USB-to-CAN converter and a network card; The USB-to-CAN device connects to the intelligent sensing system control unit via a USB interface and is responsible for converting the control unit’s control commands to the subsystem into CAN signals. The network card connects to the camera and LiDAR, enabling the intelligent sensing system control unit to obtain sensor data in real time.

4. The intelligent sensing system for a loader according to claim 2, characterized in that: The base includes a CAN communication unit for communication with the intelligent sensing system control unit, a motor control unit, and a calibration parameter storage unit; The CAN communication unit is responsible for receiving commands from the intelligent sensing system control unit and feeding back the robotic arm's motion status to it; The motor control unit is responsible for the motion control of all motors in the robotic arm; The calibration parameter storage unit is responsible for storing the motion parameters of each motor under different poses.

5. The method for an intelligent sensing system for a loader according to claim 1, characterized in that: When the intelligent perception control unit combines the perception information of the surrounding environment at the current time t0, the current pose of the loader, and the loader's motion intention information, it can predict and evaluate the new pose lp1 of the loader at the next time t1. Based on the new pose lp1 of the loader, combined with the poses s0 of each sensing device in the intelligent sensing system, the latest poses s1 of each sensing device at the next moment can be obtained. Based on the new pose s1 of each sensing device, combined with the sensing method and field of view of each sensing device and the sensing information at the current moment, the sensing information at the next moment can be comprehensively predicted and the presence of sensing blind spots in certain areas can be assessed. If a perception blind spot is detected, the robotic arm is immediately driven to move each sensing device to a designated area for environmental observation and scanning, in order to eliminate or reduce the perception blind spot at time t1 as much as possible.

6. The method for an intelligent sensing system for a loader according to claim 5, characterized in that: When the loader is in normal operation, compared with the traditional method of fixed sensing equipment, the intelligent sensing system can drive the robotic arm to conduct a comprehensive observation of the area near the bucket and boom without blind spots, eliminating blind spots in the near field of vision. While the loader is paying attention to the surrounding environment, it drives the robotic arm to periodically scan the areas on the left and right sides of the vehicle body using the sensing devices. If a person or other object is found to suddenly enter the safe area on both sides of the loader, obstacle avoidance and collision prevention measures will be taken immediately. When there are obstacles such as walls near the loader and the height of the obstacles is limited, the mechanical arm can be extended upward to give the sensing equipment a higher and better field of vision, so that the situation behind the obstacles can be understood as early as possible and safety precautions can be taken in advance. When the loader completes the digging action and plans to empty the bucket, compared to the traditional method of fixing the sensing equipment, the intelligent sensing system can make a more accurate judgment on the full bucket rate by adjusting the observation distance and angle of the sensing equipment. Since the boom lifting of the loader during unloading can easily obstruct the forward view of the sensing equipment, the robotic arm can be driven to detect and sense the unloading point from other angles, judge the material stacking in the hopper and select the best unloading point accordingly, thereby preventing uneven material stacking or avoiding spillage caused by inaccurate unloading position. When the loader completes the unloading plan and puts away the bucket, the traditional method of fixing the sensing device makes it difficult to detect whether there is any residual material in the loader bucket. The intelligent sensing system can control the robotic arm to enable the sensing device to accurately observe the material situation in the bucket. When unmanned loaders operate in confined spaces such as tunnels, the intelligent sensing system can adjust the robotic arm's posture in real time based on the actual movable space, ensuring that the robotic arm meets detection requirements to the greatest extent possible while ensuring safety and avoiding collisions.

Images