Method for controlling unmanned loader and related device

Abstract

The invention relates to a method for controlling an unmanned loader and a related device. Method for controlling an unmanned loader, the unmanned loader comprising a plurality of sets of sensors, each set of sensors for detecting a respective primary state quantity of the unmanned loader, the plurality of sets of sensors being divided into one or more sets of sensors, each set of sensors for determining a respective secondary state quantity of the unmanned loader, the method comprises the following steps: acquiring first data detected by each group of sensors in the operation process of the unmanned loader; fusing the first data of each sensor in each group of sensors to obtain second data for representing the first-level state quantity corresponding to the group of sensors; for each sensor set, performing feature extraction on the second data of each group of sensors in the sensor set to obtain a feature vector for representing the secondary state quantity of the unmanned loader; and inputting the feature vector into a fault diagnosis model to perform fault diagnosis on the unmanned loader.

Description

Technical Field

[0001] The present disclosure relates to the field of unmanned driving technology, and more particularly, to a method for controlling an unmanned loader, an apparatus for controlling an unmanned loader, an electronic device, a computer-readable storage medium, and a computer program product. Background Art

[0002] With the rapid development of intelligent equipment technology, unmanned loaders are increasingly being used in complex working environments such as mines and ports. Unmanned loaders are highly intelligent, fully automated construction machinery that integrate advanced sensors, automated control systems, and artificial intelligence algorithms to autonomously complete material handling and loading tasks without a driver. Summary of the Invention

[0003] A brief overview of the present disclosure is provided below to provide a basic understanding of some aspects of the present disclosure. However, it should be understood that this overview is not an exhaustive overview of the present disclosure. It is not intended to identify key or important parts of the present disclosure, nor is it intended to limit the scope of the present disclosure. Its purpose is simply to present certain concepts of the present disclosure in a simplified form as a prelude to the more detailed description that will be given later.

[0004] According to a first aspect of the present disclosure, a method for controlling an unmanned loader is provided, wherein the unmanned loader includes multiple groups of sensors, each group of sensors is used to detect a corresponding first-level state quantity of the unmanned loader, the multiple groups of sensors are divided into one or more sensor sets, each sensor set is used to determine a corresponding second-level state quantity of the unmanned loader, and the method includes: obtaining first data detected by each group of sensors in the multiple groups of sensors during the operation of the unmanned loader; for each group of sensors in the multiple groups of sensors, fusing the first data of each sensor in the group of sensors to obtain second data for characterizing the first-level state quantity corresponding to the group of sensors; for each sensor set in the one or more sensor sets, performing feature extraction on the second data of each group of sensors in the sensor set to obtain a feature vector for characterizing the second-level state quantity of the unmanned loader; and inputting the feature vector into a fault diagnosis model to perform fault diagnosis on the unmanned loader.

[0005] According to a second aspect of the present disclosure, a device for controlling an unmanned loader, wherein the unmanned loader includes multiple groups of sensors, each group of sensors is used to detect a corresponding first-level state quantity of the unmanned loader, the multiple groups of sensors are divided into one or more sensor sets, and each sensor set is used to determine a corresponding second-level state quantity of the unmanned loader. The device includes: a data collection module, the data collection module is coupled to the multiple groups of sensors and is configured to obtain first data detected by each group of sensors in the multiple groups of sensors during the operation of the unmanned loader; a data processing module, the data processing module is coupled to the data collection module and is configured to, for each group of sensors in the multiple groups of sensors, fuse the first data of each sensor in the group of sensors to obtain second data for characterizing the first-level state quantity corresponding to the group of sensors; a feature extraction module, the feature extraction module is coupled to the data processing module and is configured to, for each sensor set in the one or more sensor sets, perform feature extraction on the second data of each sensor set in the sensor set to obtain a feature vector for characterizing the second-level state quantity of the unmanned loader; and a fault diagnosis module, the fault diagnosis module includes a fault diagnosis model and is coupled to the feature extraction module, and the fault diagnosis module is configured to input the feature vector into the fault diagnosis model to perform fault diagnosis on the unmanned loader.

[0006] According to a third aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory storing computer-executable instructions, which, when executed by the processor, enable the processor to execute the method for controlling an unmanned loader according to the first aspect of the present disclosure.

[0007] According to a fourth aspect of the present disclosure, a computer-readable storage medium having computer-executable instructions stored thereon is provided. When the computer-executable instructions are executed by a computer, the computer executes the method for controlling an unmanned loader according to the first aspect of the present disclosure.

[0008] According to a fifth aspect of the present disclosure, a computer program product is provided. The computer program product includes instructions. When the instructions are executed by a processor, the method for controlling an unmanned loader according to the first aspect of the present disclosure is implemented. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] The foregoing and other features and advantages of the present disclosure will become apparent from the following description of the embodiments of the present disclosure taken in conjunction with the accompanying drawings, which are incorporated herein and form a part of the specification and serve to further explain the principles of the present disclosure and enable those skilled in the art to make and use the present disclosure.

[0010] Figure 1 is a flowchart illustrating a method for controlling an unmanned loader according to some embodiments of the present disclosure;

[0011] Figure 2 is a schematic block diagram illustrating an apparatus for controlling an unmanned loader according to some embodiments of the present disclosure;

[0012] Figure 3 is a schematic block diagram illustrating a non-limiting implementation of an apparatus for controlling an unmanned loader according to some embodiments of the present disclosure;

[0013] Figure 4 is a schematic block diagram illustrating an electronic device according to some embodiments of the present disclosure;

[0014] Figure 5 is a schematic block diagram illustrating a computer system upon which embodiments of the present disclosure may be implemented.

[0015] Note that in the embodiments described below, the same reference numerals are sometimes used in common across different drawings to denote the same parts or parts having the same functions, and their repeated descriptions are omitted. In some cases, similar reference numerals and letters are used to denote similar items, so once an item is defined in one drawing, it need not be further discussed in subsequent drawings.

[0016] For ease of understanding, the positions, sizes, and ranges of various structures shown in the drawings and the like may not represent actual positions, sizes, and ranges, etc. Therefore, the present disclosure is not limited to the positions, sizes, and ranges disclosed in the drawings and the like. DETAILED DESCRIPTION

[0017] Various exemplary embodiments of the present disclosure will be described in detail below with reference to the accompanying drawings. It should be noted that unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions and numerical values ​​set forth in these embodiments do not limit the scope of the present disclosure.

[0018] The following description of at least one exemplary embodiment is merely illustrative and is not intended to limit the present disclosure, its application, or use. In other words, the structures and methods herein are presented in an exemplary manner to illustrate various embodiments of the structures and methods of the present disclosure. However, those skilled in the art will appreciate that these are merely exemplary of the disclosure that may be implemented, and are not exhaustive. Furthermore, the drawings are not necessarily drawn to scale, and some features may be exaggerated to illustrate details of specific components.

[0019] In addition, technologies, methods, and equipment known to ordinary technicians in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and equipment should be considered part of the specification.

[0020] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not limiting. Therefore, other examples of the exemplary embodiments may have different values.

[0021] Unmanned loaders often malfunction during operation. In related technologies, the fault detection method for unmanned loaders can determine whether the unmanned loader has malfunctioned based on data detected by sensors such as lidar and inertial measurement units (IMUs). However, the fault detection methods in related technologies often rely on a single sensor threshold alarm mechanism. For example, the speed data of the unmanned loader detected by lidar and IMU is compared with a preset speed threshold. Once the detected speed exceeds the preset speed threshold, it is considered that the unmanned loader has malfunctioned, and an alarm is triggered. Therefore, the data type of the fault detection method in the related technology is single and the data volume is insufficient. There is a significant lag in the real-time fusion processing of multimodal data from multiple sensors used to detect different operating states of the unmanned loader. It does not have the ability to perceive and adapt to the various states of the unmanned loader in complex environments, and the fault detection accuracy is low.

[0022] To this end, the present disclosure provides a method for controlling an unmanned loader. The method fuses multimodal data from multiple sets of sensors included in the unmanned loader and generates corresponding feature vectors based on the fused data. This method then enables fault diagnosis of the unmanned loader based on the feature vectors and a fault diagnosis model. By fusing multimodal data from multiple sets of sensors, the robustness of the fault diagnosis model can be improved. Furthermore, the fault diagnosis model can obtain richer information from the fused data, thereby improving its ability to understand complex scenarios and, in turn, increasing the accuracy of its fault diagnosis for the unmanned loader.

[0023] Below, the method for controlling an unmanned loader according to various embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. It is understood that the actual method for controlling an unmanned loader may also include other steps, but to avoid obscuring the key points of the present disclosure, these other steps are not discussed herein and are not shown in the accompanying drawings.

[0024] Additionally, terms such as "first," "second," and the like may also be used herein for reference purposes only and are not intended to be limiting. For example, the terms "first," "second," and other numerical terms referring to structures or elements do not imply a sequence or order unless the context clearly indicates otherwise.

[0025] In this document, the same or similar characters may be used to represent the same or similar variables. Therefore, once a variable is defined in a certain embodiment, it does not need to be repeatedly described in subsequent embodiments.

[0026] The unmanned loader may include multiple sensor groups, each group of sensors being used to detect a corresponding primary state variable of the unmanned loader. In some embodiments, the primary state variables may include at least one of speed, temperature, position, attitude (such as heading angle, linear velocity, angular velocity, etc.), power system parameters, hydraulic system parameters (such as boom hydraulic pressure, bucket hydraulic pressure, hydraulic motor speed, etc.), and electrical system parameters (such as oil temperature, travel motor speed, battery management system parameters, etc.). For example, g1 in the multiple sensor groups {g1, g2, ..., gm} (m is a positive integer) may be a group of sensors for detecting speed, such as a Hall effect speed sensor or a magnetoelectric speed sensor. g2 in the multiple sensor groups {g1, g2, ..., gm} may be a group of sensors for detecting temperature, such as an ambient temperature sensor or a fuel temperature sensor.

[0027] The multiple groups of sensors can also be divided into one or more sensor sets, each sensor set is used to determine a corresponding secondary state quantity of the unmanned loader. The secondary state quantity corresponding to each sensor set is associated with the primary state quantity corresponding to each group of sensors in the sensor set. For example, as described in the above example, for the multiple groups of sensors {g1, g2, ... , gm}, it can also be divided into one or more sensor sets {s1, ... , sk} (k is a positive integer). Among them, g1 and g2 can be divided into the same sensor set s1, that is, s1={g1, g2}, because the speed detected by g1 and the temperature detected by g2 can both affect the engine torque of the unmanned loader (a secondary state quantity). In other words, each group of sensors in the multiple groups of sensors can be divided into a corresponding sensor set according to the secondary state quantity associated with the primary state quantity it detects.

[0028] Figure 1 FIG. 1 is a flow chart of a method 100 (hereinafter referred to as “method 100 ”) for controlling an unmanned loader according to some embodiments of the present disclosure. Figure 1 As shown, the method 100 includes steps S102 to S108.

[0029] At step S102, first data detected by each of the plurality of sensor groups during operation of the unmanned loader is acquired. In some examples, the plurality of sensor groups may include millimeter-wave radars, lidars, ultrasonic sensors, binocular vision cameras, real-time kinematic (RTK) positioning sensors, IMU sensors, speed sensors, pressure sensors, temperature sensors, and the like.

[0030] At step S104, for each group of sensors in multiple groups of sensors, the first data of each sensor in the group of sensors is fused to obtain second data for characterizing the primary state quantity corresponding to the group of sensors.

[0031] At step S106, for each sensor set in one or more sensor sets, feature extraction is respectively performed on the second data of each group of sensors in the sensor set to obtain a feature vector for characterizing the secondary state quantity of the unmanned loader.

[0032] At step S108, the feature vector is input into the fault diagnosis model to perform fault diagnosis on the unmanned loader.

[0033] In some embodiments, the first data may include multiple data points with timestamps.

[0034] In some further embodiments, method 100 may include: before fusing the first data of each sensor in each group of sensors, the first data may be processed. For example, a wavelet transform algorithm may be used to denoise the first data, and high-frequency noise interference and abnormal data are eliminated through multi-scale decomposition and reconstruction. Additionally or alternatively, in some embodiments, method 100 may include: for each group of sensors in multiple groups of sensors, before fusing the first data of each sensor in each group of sensors, the first data of each sensor in the group of sensors may be aligned with each other according to the timestamps.

[0035] For example, assume that the sampling frequency of sensor i is different from that of sensor j. Specifically, in the time period from t1 to t3, the first data detected by sensor i is {(t1, y i 1), (t2, y i 2), (t3, y i 3)}, that is, the first data of sensor i includes three data points y i 1, y i 2, and y i 3 with timestamps t1, t2, and t3 respectively, while the first data detected by sensor j is {(t1, y j 1), (t3, y j 3)}, that is, the first data of sensor j includes two data points y j 1 and y j 3 with timestamps t1 and t3 respectively, where t1 < t2 < t3. Therefore, to align the first data of sensor i with the first data of sensor j according to the timestamps, it is necessary to supplement the data point y j2. In other words, it can be considered that compared with sensor i, the first data of sensor j is missing the data point y2 with the time stamp t2. In some examples, the linear interpolation function f(t) can be used to supplement the data point y2 with the time stamp t2 as In some examples, .

[0036] By aligning the first data of each sensor with each other through timestamps, it can be ensured that data from different sensors can be processed on the same time basis, thereby ensuring that the data fused on the same time basis is valid, improving the relevance and consistency of the data, and thus improving the accuracy and robustness of the fault diagnosis model.

[0037] In some embodiments, the second data of each group of sensors is obtained by weighted summing the first data of each sensor in the group of sensors. In some examples, the second data can be set to be set to , where n is the number of sensors in the sensor group, is the first data of the i-th sensor in the group of sensors (i=1,2,...,n), is the weighting coefficient for the i-th sensor. Here, The weight of the first data of the i-th sensor in the group of sensors in the second data can be adjusted, for example, depending on the reliability of the i-th sensor during the operation of the unmanned loader or set according to current actual needs. For multiple groups of sensors {g1, g2, ..., gm}, m second data can be obtained, that is, { }.

[0038] In some examples, for each sensor set in one or more sensor sets, feature extraction is performed on the second data of each group of sensors in the sensor set to obtain a feature vector for characterizing the secondary state quantity of the unmanned loader, which may include: for each sensor set in one or more sensor sets, feature extraction is performed on the second data of each group of sensors in the sensor set to obtain a corresponding feature vector, wherein the feature vector of each group of sensors is respectively used as a feature vector component in the feature vector of the sensor set, thereby obtaining a feature vector set corresponding to the one or more sensor sets including one or more feature vectors.

[0039] As an illustrative example of feature extraction, for example, a Fourier transform can be performed on the vibration signal of the power system of an unmanned loader to extract its frequency domain features. For another example, the navigation and positioning features of the unmanned loader can be extracted by performing statistical analysis on its path curvature and positioning variance data. This disclosure does not limit the specific algorithm for feature extraction; for example, time domain analysis methods, frequency domain analysis methods, time-frequency domain analysis methods, etc. can be used. It is understood that any feature extraction algorithm currently available or achievable in the future can be applied in the embodiments of this disclosure.

[0040] For example, according to the above example, for the sensor set s1={g1, g2}, the second data of g1 is , the second data of g2 is Therefore, the second data of g1 and g2 can be extracted to obtain the corresponding feature vector components f1 and f2, thereby obtaining the feature vector of sensor set s1. ={f1, f2}. For one or more sensor sets {s1, ..., sk}, a feature vector set F={ ,... , These k feature vectors or feature vector sets will be fed into the fault diagnosis model for fault diagnosis of the unmanned loader.

[0041] In some embodiments, fault diagnosis of an unmanned loader includes: determining third data indicating the probability of a failure of the unmanned loader based on a feature vector through a fault diagnosis model; comparing the third data with a preset fault threshold; and determining that a failure of the unmanned loader occurs in response to the third data being not less than the preset fault threshold.

[0042] For example, the third data may be determined based on each eigenvector component in the eigenvector and the weight coefficient of each eigenvector component. In some examples, the third data may be set to , where N is the feature vector of the pth sensor set in one or more sensor sets The number of eigenvector components in (corresponding to the number of sensor groups in the sensor set), is the i-th eigenvector component, for The weight coefficient of is the characteristic mapping function of the fault diagnosis model, and b is the bias term.

[0043] In some further embodiments, the weight coefficient of each feature vector component is adjusted based on the corresponding environmental conditions of the operating environment of the unmanned loader. For example, the environmental conditions may include the unmanned loader's operating conditions (such as load capacity, hill climbing, flat driving, etc.) and environmental parameters (such as temperature, humidity, dust concentration, etc.).

[0044] In this embodiment, the weight coefficient of each eigenvector component is updated along with the training of the fault diagnosis model. After the fault diagnosis model is trained and in actual fault diagnosis, the weight coefficient of each eigenvector component is still updated according to the perceived environmental state. Once the perceived environmental state triggers a weight update condition (for example, the temperature exceeds a preset temperature threshold, the dust concentration exceeds a preset dust concentration threshold, etc.), the weight of the eigenvector component corresponding to the corresponding sensor group is updated according to the established strategy. In some examples, = , where E represents the corresponding environmental state, It is an adjustment factor determined according to the corresponding environmental conditions.

[0045] For example, for the aforementioned sensor set s1, in high-temperature environments, overheating is more likely to occur. Therefore, the weight of f2 corresponding to the temperature sensor set g2 can be increased, while the weight of f1 corresponding to the speed sensor set g1 can be decreased. For another example, for another sensor set s2 = {g4, g7} corresponding to the load capacity of an unmanned loader (a secondary state variable), in dusty environments, because dust can interfere with the normal operation of optical sensors, the weight of the feature vector component corresponding to the pressure sensor set g4 can be increased, while the weight of the feature vector component corresponding to the sensor set g7, used to capture images that can detect the load capacity of the unmanned loader, can be decreased.

[0046] In this way, the adaptability and robustness of the fault diagnosis model under complex and changeable working conditions can be improved, thereby improving the accuracy and reliability of the fault diagnosis model for unmanned loaders in actual environments.

[0047] The preset fault threshold may be determined based on the previously acquired first data. In some embodiments, the preset fault threshold is determined based on the mean and / or standard deviation of the previously acquired first data. In some examples, the preset fault threshold may be set to θ=μ new +k·σ new , where μ new is the mean of the first data obtained previously, σ new is the standard deviation of the first data previously obtained, and k is the adjustment coefficient. new It can represent the central tendency of the first data; σ newIt can represent the fluctuation amplitude of the first data; k can be used to control the sensitivity of the preset fault threshold to abnormal data, and the value of k is usually within 1.5 to 3.

[0048] Additionally, in some embodiments, the preset fault threshold may be updated according to a preset update frequency. When the operating state of the unmanned loader changes (such as load increase, ambient temperature change, etc.), the newly collected first data may present different statistical characteristics. new increases (i.e. the first data increases as a whole), θ will increase synchronously to match the new baseline; if σ new increases (i.e., the fluctuation of the first data intensifies), k·σ new The preset fault threshold range will be expanded to avoid misjudging normal fluctuations as faults. In addition, μ can be re-determined at fixed time intervals or after collecting a specified number of new data points. new and σ new , to avoid the preset fault threshold becoming rigid and unable to adapt to changes in the operating status of the unmanned loader in a timely manner.

[0049] For example, assuming that θ=14 MPa after the last update, based on the first data currently collected to characterize the pressure of the unmanned loader, μ of the first data is determined by calculation. new is 12.5, σ new If Pfault is 1.2 and k is set to 2, then θ = 12.5 + 2 × 1.2 = 14.9 MPa is updated. Therefore, only when Pfault is greater than 14.9 MPa is the unmanned loader considered to have failed, not when it is greater than 14 MPa.

[0050] Therefore, the fault diagnosis model updates μ in real time. new and σ new , allowing the preset fault threshold to dynamically drift with the data distribution rather than remain fixed, always maintaining adaptability to the current state of the unmanned loader. This allows for online learning of the preset fault threshold, enabling adaptive fault diagnosis and improving fault diagnosis accuracy. In contrast, fault diagnosis models in related technologies are mostly statistical models or static machine learning models based on fixed parameters. These models are difficult to adapt to nonlinear operating conditions caused by sudden load changes and dynamic path obstacle avoidance during loader operations, and their robustness is insufficient.

[0051] In some embodiments, the fault diagnosis model includes a fault classification model and a fault prediction model. In this embodiment, the fault classification model is configured to determine whether the unmanned loader is currently faulty and the type and level of the fault currently occurring based on the feature vector, and the fault prediction model is configured to determine whether the unmanned loader will fail in the future and the type and level of the fault that will occur in the future based on the feature vector. Thus, the fault diagnosis model can not only determine whether the unmanned loader is currently faulty, but also predict whether the unmanned loader will fail in the future, and can further determine the type and level of the fault that has occurred or will occur, thereby achieving comprehensive fault diagnosis of the unmanned loader. The present disclosure does not impose any special restrictions on the actual architecture of the fault classification model and the fault prediction model, as long as it can achieve the functions described herein.

[0052] In some examples, the fault classification model is built based on a support vector machine (SVM).

[0053] As an illustrative example, the training steps of a fault classification model based on SVM may include:

[0054] During the data processing phase, sensor data from the unmanned loader, both in normal operation and in various fault states, such as temperature, pressure, and vibration data, as well as equipment operating status and fault records, are first collected. Effective features reflecting the equipment status are extracted through time-domain and frequency-domain analysis methods, and each sample is labeled with the corresponding fault type to complete data labeling. Next, the features are screened, retaining the most representative and discriminative ones to reduce computational complexity. The data is standardized using methods such as standard scores or Min-Max scaling to ensure that different features have the same scale. The preprocessed dataset is then divided into training and test sets according to a certain ratio. An appropriate SVM kernel function is then selected based on the linear or nonlinear characteristics of the data (for example, a linear kernel function is suitable for linearly separable data, while a polynomial kernel function is suitable for data with certain nonlinear characteristics).

[0055] In the model training phase, the fault classification model is trained using the training set;

[0056] After the fault classification model is trained, its classification accuracy and precision are evaluated using a test set. Model parameters or features are optimized based on the evaluation results. For example, cross-validation can be used to optimize the parameters of the fault classification model. Reinforcement learning can also be used to optimize the decision boundary value of the fault classification model to improve its accuracy and adaptability.

[0057] In some examples, the fault prediction model is built based on a dual-channel Long Short-Term Memory (LSTM) network.

[0058] As an illustrative example, the training steps for a fault prediction model based on LSTM may include:

[0059] During the data processing phase, time series data from various sensors during the operation of the unmanned loader, such as vibration data, temperature data, and pressure data, are first collected. This data is then cleaned to remove noise, process missing values, and correct abnormal data. Next, features that can characterize the health status of the unmanned loader are extracted from the processed data. This includes calculating the frequency domain characteristics of the vibration signal and statistically analyzing the temperature and pressure trends. Feature selection methods are then used to select the most representative and discriminative features. The extracted features are then standardized or normalized to eliminate the influence of dimensional differences between different features. Since unmanned loader fault prediction generally requires considering the time series characteristics of the data, the processed features need to be reasonably segmented according to time windows to form an ordered input sequence and corresponding target output (such as the probability of fault occurrence or specific fault type).

[0060] During the model building phase, an LSTM network structure is constructed, and the appropriate number of hidden layers, number of neurons, and input and output dimensions are designed based on the data characteristics and prediction task requirements. The LSTM network, through its unique gating mechanism (input gate, forget gate, and output gate), can effectively capture long-term dependencies in time series data, thereby better learning how the device's operating status changes over time.

[0061] During the model training phase, the training set is used to train the model, and the validation set is used to adjust the hyperparameters. After the model training is completed, the real-time collected sensor data is input into the trained fault prediction model to obtain the fault prediction results, and the performance of the fault prediction model is verified through relevant evaluation indicators of classification or regression. Finally, a fault prediction model is established that can accurately predict the failure risk of unmanned loaders and provide early warning of potential failures.

[0062] In some embodiments, during the operation of the fault diagnosis model, the predicted results are compared with the actual fault conditions of the unmanned loader's operating state, and the error between the two is calculated. Common error indicators include mean square error and mean absolute error, and the parameters in the fault diagnosis model are adjusted based on the calculated error. For example, the weights and biases of the fault diagnosis model are updated along the gradient direction of the error so that the error gradually decreases. In this process, the learning rate is a key parameter that determines the step size of each parameter update. Therefore, an adaptive learning rate algorithm (such as but not limited to AdaGrad) can be used to automatically adjust the learning rate according to the parameter update, thereby improving the convergence speed and stability of the fault diagnosis model to adapt to the new data distribution.

[0063] In some embodiments, method 100 may include: after inputting the feature vector into the fault diagnosis model, in response to the fault classification model determining that the unmanned loader is currently faulty, determining a corresponding path planning strategy for the unmanned loader based on the level of the current fault determined by the fault classification model. In this embodiment, determining the corresponding path planning strategy for the unmanned loader may, for example, include determining the corresponding path planning strategy based on the level of the current fault and a preset matrix used to describe a mapping relationship between the level of the current fault and the path planning strategy.

[0064] For non-limiting illustrative purposes, a preset matrix as shown in Table 1 below may be established, for example.

[0065] Table 1

[0066] The mental constraint vector for path planning is C=[v_max, d_safe, ρ_reroute, δ_deviation] T , where v_max represents the maximum allowed speed (m / s), d_safe represents the minimum safe distance, ρ_reroute represents the path replanning trigger frequency (times / minute), and δ_deviation represents the allowed path deviation (as a percentage of the original path length).

[0067] When the preset matrix outputs an infeasible solution, you can perform at least one of the following operations: reduce ρ_reroute to allow fewer replanning attempts; relax δ_deviation to increase the maximum deviation by 10% to 15%; gradually reduce v_max with a step size of 0.2 m / s each time.

[0068] This preset matrix can be used to quantify the impact of faults on the performance of the unmanned loader and dynamically generate path planning constraints that match the fault level, thereby ensuring that the unmanned loader can still complete the task safely and efficiently in the fault state.

[0069] In some embodiments, method 100 may include: after inputting the feature vector into the fault diagnosis model, determining that the unmanned loader currently has a fault in response to the fault classification model, and determining a solution strategy for solving the fault based on the type and level of the current fault determined by the fault classification model.

[0070] The solution strategy can be stored in a local database or cloud corresponding to the type and level of the fault, or the relevant solution strategy can be searched from the Internet according to the type and level of the fault.

[0071] In some embodiments, method 100 may include: after inputting the feature vector into the fault diagnosis model, in response to the fault prediction model determining that the unmanned loader will fail in the future, determining a solution strategy for resolving the failure based on the type and level of the failure determined by the fault prediction model. For example, when it is determined that the unmanned loader will have a minor failure in the future, the solution strategy may include remotely controlling the unmanned loader and adjusting the operating parameters of corresponding components to achieve self-healing of the failure. For another example, when it is determined that the unmanned loader will have a serious failure in the future, the solution strategy may include promptly arranging for maintenance personnel to perform on-site repairs and providing detailed maintenance guidance.

[0072] Therefore, the unmanned loader using the embodiments of the present disclosure can autonomously detect, predict, and report faults and further adjust its own operating status (for example, the path and speed during operation, etc.) in a timely manner according to the faults to ensure the completion of the loading task and can also solve faults to a certain extent and achieve self-healing of faults.

[0073] In some embodiments, the method 100 may include displaying the results output by the fault diagnosis model on a user interface, so that an operator can directly obtain information related to the fault of the unmanned loader on the user interface, thereby facilitating the operator to promptly control and maintain the unmanned loader.

[0074] In some embodiments, method 100 may include automatically switching a control mode of the unmanned loader based on a fault level determined by a fault diagnosis model, wherein the unmanned loader typically has an unmanned mode, a manned mode, a remote operation mode, and a fault mode. Switching the control mode of the unmanned loader based on the fault level may enable the unmanned loader to promptly adopt an appropriate control mode for the fault.

[0075] The present disclosure also provides a device for controlling an unmanned loader. Figure 2 , which shows a schematic block diagram of a device 200 for controlling an unmanned loader (hereinafter referred to as “device 200 ”) according to some embodiments of the present disclosure.

[0076] like Figure 2 As shown, the unmanned loader 300 includes multiple sensor groups 302, each sensor group is used to detect a corresponding primary state quantity of the unmanned loader 300. The multiple sensor groups 302 can be divided into one or more sensor sets, each sensor set is used to determine a corresponding secondary state quantity of the unmanned loader 300.

[0077] Continue to refer Figure 2 The device 200 includes a data collection module 202 , a data processing module 204 , a feature extraction module 206 and a fault diagnosis module 208 .

[0078] The data collection module 202 is coupled to the plurality of sensor groups 302 and is configured to obtain first data detected by each sensor group in the plurality of sensor groups 302 during operation of the unmanned loader 300 .

[0079] The data processing module 204 is coupled to the data collection module 202 and is configured to, for each sensor group in the plurality of sensor groups 302, fuse the first data of each sensor in the group to obtain second data representing the primary state quantity corresponding to the sensor group. For example, the data collection module 202 may transmit the first data of each sensor group to the data processing module 204, so that the data processing module 204 can perform data fusion based on the received first data of each sensor group.

[0080] The feature extraction module 206 is coupled to the data processing module 204 and is configured to perform feature extraction on the second data of each group of sensors in each sensor set in the one or more sensor sets to obtain a feature vector for characterizing the secondary state quantity of the unmanned loader 300. For example, the data processing module 204 may transmit the second data obtained by fusing the first data to the feature extraction module 206, so that the feature extraction module 206 can perform feature extraction on the second data of each group of sensors in each sensor set to obtain a feature vector.

[0081] The fault diagnosis module 208 includes a fault diagnosis model and is coupled to the feature extraction module 206. The fault diagnosis module 208 is configured to input the feature vector into the fault diagnosis model to perform fault diagnosis on the unmanned loader 300. For example, the feature extraction module 206 may transmit the feature vector to the fault diagnosis module 208, so that the fault diagnosis module 208 may input the feature vector into the fault diagnosis model to perform fault diagnosis on the unmanned loader 300.

[0082] In some embodiments, data transmission between the various modules in device 200 is configured to be conducted via encrypted communication. For example, end-to-end encryption can be used to implement data transmission between the modules, ensuring that only the module with the correct key can decrypt and retrieve useful information. A combination of symmetric and asymmetric encryption algorithms can be used to encrypt data between the multiple sensor groups and data collection module 202 to ensure data security during transmission. For another example, the principle of least privilege can be applied to certain important data, granting only the minimum access rights required to perform a specific task and clearly defining the access rights of each module (such as the type of data that can be accessed and the conditions under which data can be accessed).

[0083] In some embodiments, as Figure 2 As shown, the apparatus 200 may include a user interface 210 coupled to the fault diagnosis module 208. The user interface 210 is configured to display results output by the fault diagnosis model.

[0084] The various embodiments of the modules in the apparatus 200 may refer to the description of the corresponding various embodiments of the method 100 above, and will not be repeated here.

[0085] Figure 3 A schematic block diagram of a non-limiting implementation 200 ′ of an apparatus 200 for controlling an unmanned loader according to some embodiments of the present disclosure is shown.

[0086] like Figure 3 As shown, the device 200' may include a data processing and feature extraction module, a safety control unit, a control output module and a human-machine interface.

[0087] For example, the data collection module 202, the data processing module 204 and the feature extraction module 206 can be implemented in a data processing and feature extraction module. The data processing and feature extraction module may include a data cleaning unit for denoising and filling data, a multi-source data fusion unit for fusing data from multiple sensors, and a multi-source data fusion unit for fusing data from multiple sensors. Figure 3 The illustrated device includes a spatiotemporal alignment processing unit for time alignment of multi-source data from multiple sensors, such as a millimeter-wave radar, a lidar, a pressure sensor, and a temperature sensor, as well as a feature extraction unit. Various embodiments of the data processing and feature extraction module can refer to the aforementioned various embodiments of the data collection module 202, the data processing module 204, and the feature extraction module 206 in the device 200.

[0088] For example, the fault diagnosis module 208 may be implemented in a safety control unit. The safety control unit includes a fault diagnosis module and an adaptive decision module. Various embodiments of the safety control unit may refer to the aforementioned various embodiments of the fault diagnosis module 208 in the apparatus 200 .

[0089] The human-machine interface may refer to the aforementioned various embodiments of the user interface 210 in the apparatus 200 .

[0090] After the safety control unit determines the fault resolution strategy or path planning strategy of the unmanned loader, it can control the corresponding sensors on the unmanned loader (for example, Figure 3 The millimeter-wave radar, lidar, pressure sensor, and temperature sensor shown in the figure are used to adjust the sensor's detection method and the corresponding execution unit (for example, engine, boom, bucket, etc.) to execute fault resolution strategy or path planning strategy.

[0091] The present disclosure also provides an electronic device, which may include: a processor; and a memory storing computer-executable instructions, which, when executed by the processor, enable the processor to execute the method for controlling an unmanned loader according to any of the aforementioned embodiments.

[0092] refer to Figure 4 , which shows a schematic block diagram of an electronic device 400 according to some embodiments of the present disclosure. Figure 4 As shown, electronic device 400 includes a processor 402 and a memory 404 storing computer-executable instructions. When executed by processor 402, the computer-executable instructions cause processor 402 to perform method 100 according to any of the aforementioned embodiments. Processor 402 may be, for example, a central processing unit (CPU) of electronic device 400. Processor 402 may be any type of general-purpose processor, or may be a processor specifically designed for controlling an unmanned loader, such as an application-specific integrated circuit ("ASIC"). Memory 404 may be coupled to processor 402 and may include various computer-readable media accessible by processor 402. In various embodiments, memory 404, as described herein, may include volatile and non-volatile media, removable and non-removable media. For example, memory 404 may include any combination of random access memory ("RAM"), dynamic RAM ("DRAM"), static RAM ("SRAM"), read-only memory ("ROM"), flash memory, cache memory, and / or any other type of non-transitory computer-readable media. The memory 404 may store instructions that, when executed by the processor 402 , enable the processor 402 to perform the method 100 according to any of the aforementioned embodiments of the present disclosure.

[0093] The electronic device 400 is configured to execute the method 100 described in any of the aforementioned embodiments, and therefore reference may be made to the aforementioned descriptions of the various embodiments of the method 100 , which will not be repeated here.

[0094] The present disclosure also provides a computer-readable storage medium having computer-executable instructions stored thereon. When the computer-executable instructions are executed by a computer, the computer is caused to execute the method for controlling an unmanned loader according to any of the aforementioned embodiments of the present disclosure.

[0095] The present disclosure also provides a computer program product comprising instructions that, when executed by a processor, implement the method for controlling an unmanned loader according to any of the aforementioned embodiments of the present disclosure. The instructions may be any set of instructions to be directly executed by one or more processors, such as machine code, or any set of instructions to be indirectly executed, such as a script. The instructions may be stored in an object code format for direct processing by one or more processors, or in any other computer language, including scripts or collections of independent source code modules that are interpreted on demand or compiled in advance.

[0096] Figure 5A schematic block diagram of a computer system 500 on which embodiments of the present disclosure may be implemented is shown. Computer system 500 includes a bus 502 or other communication mechanism for communicating information, and a processing device 504 coupled to bus 502 for processing information. Computer system 500 also includes a memory 506 coupled to bus 502 for storing instructions to be executed by processing device 504. Memory 506 may be a random access memory (RAM) or other dynamic storage device. Memory 506 may also be used to store temporary variables or other intermediate information during the execution of instructions to be executed by processing device 504. Computer system 500 also includes a read-only memory (ROM) 508 or other static storage device coupled to bus 502 for storing static information and instructions for processing device 504. A storage device 510, such as a magnetic disk or optical disk, is provided and coupled to bus 502 for storing information and instructions. Computer system 500 may be coupled via bus 502 to output devices 512 for providing output to a user, such as, but not limited to, a display (such as a cathode ray tube (CRT) or liquid crystal display (LCD)), speakers, and the like. Input devices 514, such as a keyboard, mouse, microphone, and the like, are coupled to bus 502 for communicating information and command selections to processing device 504. Computer system 500 may perform embodiments of the present disclosure. Consistent with certain implementations of the present disclosure, results are provided by computer system 500 in response to processing device 504 executing one or more sequences of one or more instructions contained in memory 506. Such instructions may be read into memory 506 from another computer-readable medium, such as storage device 510. Execution of the sequences of instructions contained in memory 506 causes processing device 504 to perform the methods described herein. Alternatively, hardwired circuitry may be used in place of or in combination with software instructions to implement the present teachings. Therefore, implementations of the present disclosure are not limited to any specific combination of hardware circuitry and software. In various embodiments, computer system 500 can be connected to one or more other computer systems like computer system 500 across a network via network interface 516 to form a networked system. The network can include a private network or a public network such as the Internet. In a networked system, one or more computer systems can store data and supply data to other computer systems. As used herein, the term "computer-readable medium" refers to any medium that participates in providing instructions to processing device 504 for execution. Such media can take many forms, including but not limited to non-volatile media, volatile media, and transmission media. Non-volatile media include, for example, optical or magnetic disks such as storage device 510. Volatile media include dynamic memory such as memory 506. Transmission media include coaxial cables, copper wire, and optical fibers, including the wiring comprising bus 502.Common forms of computer-readable media or computer program products include, for example, floppy disks, flexible disks, hard disks, magnetic tape, or any other magnetic medium, CD-ROMs, digital video disks (DVDs), Blu-ray discs, any other optical medium, thumb drives, memory cards, RAM, PROMs and EPROMs, flash EPROMs, any other memory chips or cartridges, or any other tangible medium from which a computer can read. Various forms of computer-readable media may be involved in carrying one or more sequences of one or more instructions to processing device 504 for execution. For example, the instructions may initially be carried on a diskette of a remote computer. The remote computer may load the instructions into its dynamic memory and send the instructions over a telephone line using a modem. A modem local to computer system 500 may receive the data over the telephone line and convert the data into an infrared signal using an infrared transmitter. An infrared detector coupled to bus 502 may receive the data carried in the infrared signal and place the data on bus 502. Bus 502 carries the data to memory 506, from which processing device 504 retrieves the instructions and executes them. Optionally, the instructions received by memory 506 may be stored on storage device 510 either before or after execution by processing device 504 .

[0097] According to various embodiments, instructions configured to be executed by a processing device to perform a method are stored on a computer-readable medium. A computer-readable medium may be a device that stores digital information. For example, a computer-readable medium includes a compact disk read-only memory (CD-ROM) as is known in the art for storing software. The computer-readable medium is accessed by a processor adapted to execute the instructions configured to be executed.

[0098] The foregoing description describes one or more exemplary embodiments of the present disclosure. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in an order different from that described in the exemplary embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0099] The systems, devices, modules or units described in the above embodiments may be implemented by computer chips or entities, or by products having certain functions. A typical implementation device is a server system. Of course, the present disclosure does not exclude that with the future development of computer technology, the computer that implements the functions of the above embodiments may be, for example, a personal computer, a laptop computer, an in-vehicle human-computer interaction device, a cellular phone, a camera phone, a smart phone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0100] Although one or more embodiments of the present disclosure provide method operation steps as described in the embodiments or flowcharts, more or fewer operation steps may be included based on conventional or non-inventive means. The order of steps listed in the embodiments is only one way of executing the steps among many steps and does not represent the only execution order. When an actual device or terminal product is executed, the method can be executed sequentially or in parallel according to the embodiments or the accompanying drawings (for example, in a parallel processor or multi-threaded processing environment, or even a distributed data processing environment).

[0101] The terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, product, or apparatus that includes a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, product, or apparatus. Without further limitation, the presence of additional identical or equivalent elements in a process, method, product, or apparatus that includes the elements is not precluded. For example, if words such as "first," "second," etc. are used to indicate names, they do not imply any particular order.

[0102] For the convenience of description, the above devices are described in terms of functions divided into various modules. Of course, when implementing one or more embodiments of the present disclosure, the functions of each module can be implemented in the same or multiple software and / or hardware, or the module that implements the same function can be implemented by a combination of multiple sub-modules or sub-units, etc. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0103] The present disclosure is described with reference to flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present disclosure. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device produce a device for implementing the functions specified in one or more processes in the flowcharts and / or one or more blocks in the block diagrams.

[0104] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce a manufactured product including an instruction device that implements the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0105] These computer program instructions may also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, whereby the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more processes in the flowchart and / or one or more boxes in the block diagram.

[0106] Those skilled in the art will appreciate that one or more embodiments of the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, one or more embodiments of the present disclosure may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0107] One or more embodiments of the present disclosure may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. One or more embodiments of the present disclosure may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0108] The same or similar parts between the various embodiments of the present disclosure can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment. In the description of the present disclosure, the reference terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" and the like mean that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In the present disclosure, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in an appropriate manner. In addition, those skilled in the art can combine and combine the different embodiments or examples described in the present disclosure and the features of the different embodiments or examples without contradiction.

[0109] In addition, when used in this disclosure, the words "herein," "above," "below," "hereunder," "supra," and words of similar meaning shall refer to the disclosure as a whole and not to any particular portions of the disclosure. Furthermore, unless expressly stated otherwise or otherwise understood in the context of use, conditional language used herein, such as "may," "might," "for example," "such as," and the like, is generally intended to convey that certain embodiments include, while other embodiments do not include, certain features, elements, and / or states. Thus, such conditional language is generally not intended to imply that one or more embodiments in any way require features, elements, and / or states, or whether such features, elements, and / or states are included or performed in any particular embodiment.

Claims

1. A method for controlling an unmanned loader, the unmanned loader comprising multiple sensor groups, each sensor group configured to detect a corresponding primary state quantity of the unmanned loader, the multiple sensor groups being divided into one or more sensor sets, each sensor set configured to determine a corresponding secondary state quantity of the unmanned loader, the method comprising: acquiring first data detected by each of the plurality of sensor groups during operation of the unmanned loader; For each group of sensors in the plurality of groups of sensors, first data of each sensor in the group of sensors are fused to obtain second data for representing a primary state quantity corresponding to the group of sensors; For each sensor set in the one or more sensor sets, feature extraction is performed on the second data of each group of sensors in the sensor set to obtain a feature vector for characterizing the secondary state quantity of the unmanned loader; as well as The feature vector is input into a fault diagnosis model to perform fault diagnosis on the unmanned loader.

2. The method according to claim 1, wherein The first data includes a plurality of data points with time stamps, For each sensor group in the plurality of sensor groups, before fusing the first data of each sensor in each sensor group, the first data of each sensor in the sensor group are aligned with each other according to timestamps.

3. The method according to claim 1, wherein The second data of each group of sensors is obtained by weighted summing the first data of each sensor in the group of sensors.

4. The method according to claim 1, wherein Performing fault diagnosis on the unmanned loader includes: determining, by the fault diagnosis model and based on the feature vector, third data indicating a probability of a fault occurring in the unmanned loader; comparing the third data with a preset fault threshold, the preset fault threshold being determined based on the previously acquired first data; and In response to the third data being not less than the preset fault threshold, it is determined that the unmanned loader has failed.

5. The method according to claim 4, wherein The third data is determined based on each eigenvector component in the eigenvector and a weight coefficient of each eigenvector component, wherein the weight coefficient of each eigenvector component is adjusted based on a corresponding environmental state in the operating environment of the unmanned loader.

6. The method according to claim 4, wherein: The preset fault threshold is determined based on a mean value and / or a standard deviation of the previously acquired first data.

7. The method according to claim 4, wherein: The preset fault threshold is updated according to a preset update frequency.

8. The method according to claim 1, wherein The fault diagnosis model includes a fault classification model and a fault prediction model, wherein the fault classification model is configured to determine whether the unmanned loader is currently faulty and the type and level of the current fault based on the feature vector, and the fault prediction model is configured to determine whether the unmanned loader will fail in the future and the type and level of the future fault based on the feature vector.

9. The method according to claim 8, wherein The fault classification model is established based on a support vector machine; and / or The fault prediction model is established based on a dual-channel long short-term memory network.

10. The method according to claim 8, comprising: After the feature vector is input into the fault diagnosis model, in response to the fault classification model determining that the unmanned loader currently has a fault, a corresponding path planning strategy for the unmanned loader is determined based on the level of the current fault determined by the fault classification model.

11. The method according to claim 10, wherein: Determining the corresponding path planning strategy of the unmanned loader includes: The corresponding path planning strategy is determined based on the level of the currently occurring fault and a preset matrix used to describe the mapping relationship between the level of the currently occurring fault and the path planning strategy.

12. The method according to claim 8, comprising: After the feature vector is input into the fault diagnosis model, in response to the fault prediction model determining that the unmanned loader will fail in the future, a solution strategy for resolving the failure is determined based on the type and level of the future failure determined by the fault prediction model.

13. The method according to claim 1 or 10 or 12, comprising: The results output by the fault diagnosis model are displayed on a user interface.

14. A device for controlling an unmanned loader, wherein: The unmanned loader includes multiple groups of sensors, each group of sensors is used to detect a corresponding primary state quantity of the unmanned loader, the multiple groups of sensors are divided into one or more sensor sets, each sensor set is used to determine a corresponding secondary state quantity of the unmanned loader, and the device includes: a data collection module coupled to the plurality of sensor groups and configured to obtain first data detected by each sensor group in the plurality of sensor groups during operation of the unmanned loader; a data processing module coupled to the data collection module and configured to, for each group of sensors in the plurality of groups of sensors, fuse first data of respective sensors in the group of sensors to obtain second data for representing a primary state quantity corresponding to the group of sensors; a feature extraction module, coupled to the data processing module and configured to perform feature extraction on the second data of each group of sensors in each of the one or more sensor sets to obtain a feature vector for characterizing a secondary state quantity of the unmanned loader; A fault diagnosis module includes a fault diagnosis model and is coupled to the feature extraction module. The fault diagnosis module is configured to input the feature vector into the fault diagnosis model to perform fault diagnosis on the unmanned loader.

15. The apparatus according to claim 14, further comprising: A user interface is coupled to the fault diagnosis module and is configured to display a result output by the fault diagnosis model.

16. An electronic device comprising: processor; as well as A memory storing computer-executable instructions, which, when executed by the processor, cause the processor to perform the method for controlling an unmanned loader according to any one of claims 1 to 13. 17 . A computer-readable storage medium having computer-executable instructions stored thereon, which, when executed by a computer, cause the computer to execute the method for controlling an unmanned loader according to claim 1 . 18 . A computer program product comprising instructions, which, when executed by a processor, implement the method for controlling an unmanned loader according to claim 1 .

Images