An electric mine truck adaptive driving mode switching control method and device

Abstract

The application discloses a kind of electric mine truck adaptive driving mode switching control method and device, belong to new energy vehicle intelligent control technical field.By obtaining multi-source sensing signal and SOC, filter extraction slope, flatness, road type and adhesion coefficient characteristics, road condition category and confidence are output by D-S evidence theory fusion decision;When confidence is over threshold and road condition is not consistent with current mode, safety constraint check is carried out: if long downhill and SOC is greater than high power threshold, it is prohibited to cut in, recycle to zero and alarm, and wait for fall to remove;If, target parameter group is called, torque response slope, maximum speed limit and recycle intensity are configured transition time, linear slope is smoothly changed and output to motor controller, vehicle controller and BMS execution;Cyclic execution.The application solves the problem of power impact and overcharge caused by switching under complex mine road conditions, and realizes safe and smooth adaptive switching.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for new energy vehicles, and in particular to an adaptive driving mode switching control method, device, equipment, and storage medium for electric mining trucks. Background Technology

[0002] The internal transportation roads in open-pit coal mines are characterized by highly dynamic changes and extreme working conditions. Field surveys of multiple open-pit mines revealed that their roads primarily encompass the following typical scenarios: long-distance, heavy-load uphill sections, where mine pits can reach depths of over 100 meters, and the gradient of transport ramps connecting different levels is typically 6%–12%, placing extremely high demands on the continuous output capacity of the power system; long-distance downhill sections, where relying solely on mechanical brakes during heavy-load descents is highly susceptible to brake failure due to heat fade; frost-prone sections, where the roadbed bearing capacity drops sharply during the spring thawing period, causing the road surface to deform in a wavy pattern, making vehicles prone to getting stuck or axle damage; muddy and waterlogged sections, where tire adhesion decreases significantly after heavy rains during the rainy season, and there is a risk of high-pressure systems wading through water; and icy and snow-covered sections, where the road surface freezes in winter, significantly increasing the risk of vehicle skidding.

[0003] Currently, major domestic and international electric mining truck manufacturers primarily employ three technical solutions for driving mode control. Solution 1 is fixed parameter control, where the vehicle controller uses a fixed set of power output parameters and energy recovery strategies, which do not adjust according to road conditions. This solution cannot simultaneously address the power demands for climbing, braking safety on downhill slopes, and anti-skid requirements on wet roads, resulting in poor overall performance. Solution 2 is manual mode switching, allowing the driver to manually select the appropriate mode via a physical knob or touchscreen. This solution suffers from drawbacks such as response lag, driver misjudgment or omissions, poor adaptability to complex road conditions, and increased driver workload. Solution 3 is automatic switching based on a single sensor, utilizing a slope sensor to switch between uphill and downhill modes. However, its recognition dimension is limited, failing to distinguish the different torque response requirements of frost-covered slopes versus normal slopes, and also unable to identify low-adhesion, non-slope special road conditions such as snow accumulation or water accumulation.

[0004] Therefore, there is an urgent need for a control method that can integrate multi-source road condition information, automatically identify complex road scenarios, and achieve smooth switching of driving modes, so as to improve the power, safety and adaptability of electric mining trucks under harsh road conditions in open-pit mines. Summary of the Invention

[0005] The present invention aims to at least partially solve one of the technical problems in the related art.

[0006] To address this issue, this invention proposes an adaptive driving mode switching control method for electric mining trucks. By acquiring multi-source sensor signals and the battery's State of Charge (SOC), the method filters and extracts features such as slope, smoothness, road surface type, and adhesion coefficient. These features are then fused using DS evidence theory to output the road condition category and confidence level. When the confidence level exceeds a threshold and the road condition is inconsistent with the current mode, a safety constraint check is performed based on the SOC. If it's a long downhill slope and the SOC is greater than or equal to a high battery threshold, switching is prohibited, recovery is reset to zero, and an alarm is triggered, which is lifted upon descent. If the check passes, the target parameter set is invoked, configuring the torque response slope, maximum speed limit, and recovery intensity with a transition time, and outputting a smooth, gradual transition via a linear slope. This process is repeated cyclically at a preset period. This invention solves the problem of power shock and overcharging during switching in complex mining area road conditions, achieving safe, smooth, and adaptive optimization of mode switching.

[0007] Another objective of this invention is to provide an adaptive driving mode switching control device for electric mining trucks.

[0008] The third objective of this invention is to provide a computer device.

[0009] A fourth objective of this invention is to provide a non-transitory computer-readable storage medium.

[0010] To achieve the above objectives, the present invention provides an adaptive driving mode switching control method for electric mining trucks, comprising: The system acquires multi-source sensor signals and battery state of charge (SOC) of electric mining trucks. After filtering, it extracts features such as slope, smoothness, road surface type and adhesion coefficient. The features are then input into the DS evidence theory for fusion and decision-making, and the current road condition category and confidence level are output. When the confidence level exceeds the threshold and the road condition category is inconsistent with the current mode, a safety constraint check is performed based on the battery state of charge (SOC): if the road condition is a long downhill and the SOC is greater than or equal to the high charge threshold, then switching to the long downhill mode is prohibited, energy recovery is reset to zero and an alarm is triggered. Once the SOC falls below the threshold, the prohibition is lifted and the mode can be switched to. If the check passes, the target control parameter group corresponding to the road condition is called. The torque response slope, maximum vehicle speed limit, and energy recovery intensity in the target control parameter group are configured with transition times, and after being smoothly transitioned to the target value through a linear ramp function, they are output to the motor controller, vehicle controller, and battery management system for execution. The above steps are executed repeatedly at a preset cycle.

[0011] An adaptive driving mode switching control method for electric mining trucks according to an embodiment of the present invention may also have the following additional technical features: In one embodiment of the present invention, acquiring multi-source sensor signals from an electric mining truck includes: At a sampling frequency of no less than 50Hz, the longitudinal acceleration, pitch angle and vertical acceleration signals output by the six-axis IMU, the road surface depth image and visible light image signal output by the binocular stereo camera, the wheel speed signals of each drive wheel output by the wheel slip rate sensor, and the road surface temperature and humidity signals output by the road surface temperature and humidity sensor are collected respectively. The system receives the battery state of charge (SOC) output by the battery management system (BMS) via the vehicle's CAN bus, as well as the current vehicle speed and motor temperature fed back by the motor controller (MCU), which together with these signals constitute a multi-source sensing signal.

[0012] In one embodiment of the present invention, the features of slope, smoothness, road surface type, and adhesion coefficient are extracted after filtering, including: The multi-source sensor signals are input into the signal acquisition and conditioning unit, and after Kalman filtering and noise reduction, a pre-processed multi-dimensional sensor dataset is generated. The real-time slope value is calculated based on the pitch angle in the multidimensional sensor dataset, and will remain for more than 3 seconds. The slope value is used as a candidate for slope road condition and output as a slope feature. The variance is calculated based on the vertical acceleration in the multidimensional sensor dataset, and the variance value is greater than... It is used as a candidate for road surface roughness and output as a feature of road surface smoothness; The depth images and visible light images in the multidimensional sensing dataset are input into a lightweight convolutional neural network. After classification and reasoning, the network outputs visual confidence vectors for four types of road surfaces: water accumulation, mud, snow and ice, and normal road surfaces, which are used as road surface type features. The slip ratio is calculated based on the wheel speed of each drive wheel in the multidimensional sensor dataset, and the adhesion coefficient feature is derived by combining the road surface temperature and humidity.

[0013] In one embodiment of the present invention, the various features are input into the DS evidence theory and fused into a decision, outputting the current road condition category and confidence level, including: Slope characteristics, road surface smoothness characteristics, road surface type characteristics, and adhesion coefficient characteristics are each constructed as an independent evidence body, and a corresponding basic confidence assignment function is assigned to each evidence body. The basic confidence assignment function of each piece of evidence is input into the DS evidence theory fusion center, and the Dempster synthesis rule is used for fusion to output the comprehensive confidence of each candidate road condition category. Select the one with the highest overall confidence level that exceeds the preset threshold. The candidate road condition categories are output as the current road condition category, and the corresponding comprehensive confidence level is output as the confidence level of the corresponding decision result.

[0014] In one embodiment of the present invention, performing a safety constraint check based on the battery state of charge (SOC) includes: When the current road condition is a long downhill slope, the current SOC value output in real time by the Battery Management System (BMS) is read, and the current SOC value is compared with a preset high charge threshold. Compare; If the current SOC value is greater than or equal to Then the mode decision unit generates a prohibition command to prohibit entering the long downhill mode and a voice alarm command that the battery is too high and the electric braking will weaken. At the same time, it forces the target energy recovery intensity to be set to zero and outputs it to the battery management system (BMS). The current SOC value is continuously read at a preset monitoring period. When the current SOC value falls below a certain level... When the time comes, the mode decision unit automatically cancels the prohibition command, restores the strong energy recovery setting, determines the current road condition category as long downhill mode, and calls the corresponding target control parameter group.

[0015] In one embodiment of the present invention, the process of smoothly transitioning to the target value via a linear ramp function includes: Read the current control parameter set in the current active mode, which includes the current torque response slope, the current maximum vehicle speed limit, and the current energy recovery intensity. The target torque response slope, target maximum vehicle speed limit, and target energy recovery intensity are read from the target control parameter set, and a preset transition time is configured for each control parameter. Preset transition time Set according to parameter type Within seconds; The current value, the target value, and their respective preset transition times are set. Substitute into the linear ramp function:

[0016] in The current value, For the target value, The preset transition time, The time is counted from the start of the switchover and ; The control parameters are calculated in real time at the current moment according to a preset calculation cycle. Value, when When all control parameters reach their corresponding target values, a smooth transition is achieved.

[0017] In one embodiment of the present invention, in the DS evidence theory fusion decision, when the current road condition category is frost-prone road, a multi-feature joint determination method is adopted, including: Extract the rut depth and lateral unevenness values ​​of the road surface ahead from the depth image output by the binocular stereo camera; Extract the peak-to-peak value of the fluctuation from the vertical acceleration signal output by the six-axis IMU; When the rut depth value is greater than Lateral unevenness value greater than And the peak-to-peak value of the vertical acceleration fluctuation is greater than At that time, the above three judgment conditions are input into the DS Evidence Theory Fusion Center as independent evidence of the frost-prone road. The DS Evidence Theory Fusion Center performs Dempster synthesis rule fusion on the three evidence bodies and outputs a comprehensive confidence score for the sluice gate category. When the comprehensive confidence score exceeds a preset threshold... At that time, the current road condition category output is "flooded road".

[0018] To achieve the above objectives, another aspect of the present invention provides an adaptive driving mode switching control device for electric mining trucks, comprising: The fusion recognition module is used to acquire multi-source sensor signals and battery state of charge (SOC) of electric mining trucks. After filtering, it extracts features such as slope, flatness, road surface type and adhesion coefficient. The features are then input into the DS evidence theory for fusion judgment, and the current road condition category and confidence level are output. The safety decision module is used to perform a safety constraint check based on the battery state of charge (SOC) when the confidence level exceeds the threshold and the road condition category is inconsistent with the current mode: if the road condition is a long downhill and the SOC is greater than or equal to the high charge threshold, then the long downhill mode is prohibited, energy recovery is set to zero and an alarm is triggered. The prohibition is lifted and the mode is enabled after the SOC falls below the threshold. If the check passes, the target control parameter group corresponding to the road condition is called. The smooth execution module is used to configure the torque response slope, maximum vehicle speed limit and energy recovery intensity in the target control parameter group with transition time, and then smoothly transition them to the target value through a linear ramp function, and output them to the motor controller, vehicle controller and battery management system for execution respectively. The loop control module is used to execute the above steps in a preset cycle.

[0019] This invention discloses an adaptive driving mode switching control method and device for electric mining trucks. By acquiring multi-source sensor signals and the battery's state of charge (SOC), filtering and extracting multi-dimensional road condition features, and then inputting these features into a DS evidence theory fusion decision, the method outputs the current road condition category and confidence level. When the confidence level exceeds a threshold and the road condition does not match the current mode, a safety constraint check is performed based on the SOC. Only when the SOC meets the conditions can the corresponding mode be switched and the target control parameter group be invoked. Each control parameter is configured with a transition time and smoothly transitioned to the target value using a linear ramp function before being output for execution, and the process iterates cyclically at a preset period. This method achieves a closed-loop adaptive control throughout the entire process, from road condition perception, fusion decision, safety verification to smooth parameter transition. It effectively solves the problems of fixed parameters being unsuitable for complex mining road conditions, manual switching lag, and insufficient recognition dimensions of a single sensor. This significantly improves the power and safety of electric mining trucks under harsh road conditions, ensuring the smoothness of the mode switching process and the safety and reliability of the battery system.

[0020] To achieve the above objectives, a third aspect of this application provides a computer device, including a processor and a memory; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, for implementing an adaptive driving mode switching control method for electric mining trucks as described in the first aspect embodiment.

[0021] To achieve the above objectives, a fourth aspect of this application provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements an adaptive driving mode switching control method for electric mining trucks as described in the first aspect embodiment.

[0022] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0023] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of an adaptive driving mode switching control method for an electric mining truck according to an embodiment of the present invention; Figure 2 This is a system hardware architecture diagram according to an embodiment of the present invention; Figure 3 This is a flowchart of the road condition recognition and mode switching control according to an embodiment of the present invention; Figure 4 This is a flowchart of the SOC adaptive logic under the long downhill mode according to an embodiment of the present invention; Figure 5This is a schematic diagram of the structure of an adaptive driving mode switching control device for an electric mining truck according to an embodiment of the present invention; Figure 6 It is a computer device according to an embodiment of the present invention. Detailed Implementation

[0024] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0025] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0026] The following description, with reference to the accompanying drawings, describes an adaptive driving mode switching control method, apparatus, device, and storage medium for electric mining trucks according to embodiments of the present invention.

[0027] Example 1 like Figure 1 As shown, the adaptive driving mode switching control method for electric mining trucks of the present invention includes the following steps: S1 acquires multi-source sensor signals and battery state of charge (SOC) of the electric mining truck. After filtering, it extracts features such as slope, smoothness, road surface type, and adhesion coefficient. It then inputs each feature into the DS evidence theory for fusion and decision-making, and outputs the current road condition category and confidence level.

[0028] In one embodiment of the present invention, a road condition recognition unit is first used, employing a multi-sensor fusion algorithm, specifically including: S101, Data Acquisition and Preprocessing. Each sensor acquires signals at a frequency of no less than 50Hz, which are then simultaneously input to the road condition recognition unit after Kalman filtering to remove noise.

[0029] S102, Feature Extraction. Slope Features: Real-time slope values ​​are calculated from the IMU pitch angle; slopes exceeding ±5° for more than 3 seconds are considered candidate road conditions. Road Surface Smoothness Features: Variance is calculated from the IMU vertical acceleration; slopes greater than 0.3g are considered candidate roads prone to frost heave. Road Surface Type Features: Classification is performed from binocular visual images using a lightweight convolutional neural network, outputting confidence scores for four categories: water accumulation, mud, snow / ice, and normal. Adhesion Coefficient Features: Slip rate is calculated from wheel speed sensors and combined with data from road surface temperature and humidity sensors for comprehensive judgment.

[0030] Then, a fusion decision is made. That is, in S103, the above features are fused using the DS evidence theory to output the current road condition category and its confidence level.

[0031] It is worth mentioning that the multi-feature fusion recognition of frost heave and snow accumulation modes includes: Frost heave road recognition uses a combination of visual features and IMU features: when the binocular camera detects road surface wave deformation (rut depth > 10cm, lateral unevenness > 8%), and the peak-to-peak value of the IMU vertical acceleration fluctuation is greater than 0.4g, the frost heave road confidence score is fused out; snow accumulation road recognition uses a combination of visual features and temperature features: when the temperature sensor is below 2℃, the visual recognition road surface is white-covered and the texture is blurred, the snow accumulation mode confidence score is fused out.

[0032] S2, when the confidence level exceeds the threshold and the road condition category is inconsistent with the current mode, a safety constraint check is performed based on the battery state of charge (SOC): if the road condition is a long downhill and the SOC is greater than or equal to the high charge threshold, then the long downhill mode is prohibited, energy recovery is set to zero and an alarm is triggered. The prohibition is lifted and the mode is activated after the SOC falls below the threshold. If the check passes, the target control parameter group corresponding to the road condition is called.

[0033] In one embodiment of the present invention, a mode switching request is triggered when the confidence level is higher than a preset threshold (e.g., 0.7). This is the SOC adaptive logic for the long downhill mode.

[0034] When the battery's State of Charge (SOC) is above 90%, the electric braking capability will decrease. This system adds SOC judgment to the mode decision-making process: if the SOC is ≥ 90% and the road condition is identified as a long downhill slope, the system will automatically issue a voice alarm "Battery level too high, electric braking will weaken soon," and temporarily prohibit switching to the long downhill mode, forcibly set the energy recovery intensity to 0, and prompt the driver to use the mechanical brakes; if the SOC drops below 90%, it will automatically switch back to the long downhill mode and restore strong energy recovery.

[0035] S3, configure the torque response slope, maximum vehicle speed limit and energy recovery intensity in the target control parameter group with transition time, and after smoothing and gradually changing to the target value through a linear ramp function, output them to the motor controller, vehicle controller and battery management system for execution respectively.

[0036] In one embodiment of the present invention, the main focus is on mode switching and smooth transition, including: Safety constraint check: If the requested mode is inconsistent with the current mode, the mode decision unit first performs a safety constraint check based on the vehicle status (vehicle speed, motor temperature, battery SOC, etc.). For example, it will not switch to the traction mode when the vehicle speed is higher than 30km / h, will not switch to the heavy-load climbing mode when the motor temperature is too high, and will not switch to the long downhill mode when the SOC is higher than 90%.

[0037] Smooth Transition Execution: After passing the safety constraint check, the smooth transition unit is activated, using a ramp function to gradually change the control parameters. Specifically, for each control parameter that needs adjustment (such as torque response slope, maximum vehicle speed limit, energy recovery intensity, etc.), the system executes according to the following ramp function:

[0038] Where: P(t): real-time parameter value at time t; P_current: parameter value of the current mode; P_target: parameter value of the target mode; T_ramp: preset transition time, set within the range of 0.5 to 2 seconds depending on the parameter type (e.g., torque response slope transition time is set to 1 second, maximum speed limit transition time is set to 0.5 seconds); T: time counted from the start of switching, with a value range of 0 ≤ t ≤ Tramp.

[0039] It should be noted that this function causes the parameters to change linearly over the transition time, rather than jumping instantaneously. Taking the switch from "Normal Mode" to "Long Uphill Mode" as an example, the torque response slope decreases uniformly from 2000 Nm / s to 800 Nm / s within 1 second, and the maximum speed limit decreases uniformly from 40 km / h to 25 km / h within 0.5 seconds.

[0040] It should be noted that the transition is complete when t=T_ramp, all parameters have reached the target value, the system updates the current mode to the target mode, and completes a full mode switch.

[0041] S4, execute the above steps repeatedly at a preset cycle.

[0042] In one embodiment of the present invention, a cyclic update is performed: the entire identification and switching process is repeatedly executed at a frequency of 10Hz to achieve dynamic adaptation when road conditions change.

[0043] The method of this invention effectively solves the problems of poor adaptability of fixed parameters, lag in manual switching, and insufficient recognition dimensions of a single sensor under complex mining road conditions. It realizes adaptive mode switching and smooth parameter transition under road conditions, significantly improving the power, safety, and driving smoothness of electric mining trucks under harsh road conditions, while ensuring the safety and reliability of the battery system under saturated charge conditions.

[0044] Example 2 It is worth mentioning that this invention also provides an adaptive driving mode switching control system for electric mining trucks. While retaining the original vehicle's electronic and electrical architecture, the system adds multiple types of sensors to perceive the vehicle's own attitude and the road conditions ahead, and reuses existing signals from the BMS, MCU, and other components on the vehicle's CAN bus. Through a multi-sensor information fusion algorithm, it automatically identifies the current and forward road conditions, and makes safety decisions based on vehicle status (vehicle speed, load, battery SOC, motor temperature, etc.), achieving autonomous and smooth switching of driving modes.

[0045] It should be noted that, as Figure 2 As shown, the overall system architecture mainly includes the following units: 1) Signal acquisition unit: Receives raw signals from each installed sensor and multiplexed signals from the CAN bus, and performs filtering and synchronization processing; 2) Road condition recognition unit: Based on a multi-sensor data fusion algorithm, it outputs the current and predicted road condition categories ahead; 3) Mode Decision Unit: Selects the optimal driving mode based on road condition recognition results, vehicle status, and safety constraints; 4) Parameter Adaptive Unit: Maps the selected mode to control parameters that can be executed by VCU, MCU, and BMS; 5) Smooth transition unit: During mode switching, step-like or ramp-like parameter adjustments are used to avoid abrupt control changes.

[0046] It should be understood that this invention adds the following sensors to the existing electric mining truck: a forward-looking binocular stereo camera: installed above the windshield of the cab, used to identify the road slope, road surface type (water / mud / ice / snow), and rut depth, outputting an image + depth information signal; a millimeter-wave radar: installed in the middle of the front bumper; used to detect road undulations and obstacles; outputting a distance + speed information signal; a six-axis IMU: installed near the vehicle's center of gravity; used to measure the vehicle's longitudinal / lateral acceleration, pitch angle, and roll angle in real time; outputting an acceleration + angular velocity signal; a wheel slip ratio sensor: installed in the hub of each drive wheel; used to detect the tire-road adhesion coefficient and slippage state; outputting a wheel speed + slip ratio signal; a road surface temperature / humidity sensor: installed under the front axle; used to determine whether the road surface is icy or waterlogged; outputting a temperature + humidity signal. In addition, the system reuses existing vehicle signals via the CAN bus, including the SOC and battery temperature output by the BMS, and the motor temperature, speed, and torque fed back by the MCU.

[0047] It should be understood that the present invention presets the following six dedicated driving modes, and the key control parameters of each mode are shown in Table 1 below: Table 1

[0048] The torque response slope is set differently based on the protection targets and adhesion limits corresponding to different modes. The slope of the water / mud mode is lower than that of the long uphill mode to prioritize the anti-skid performance of low-adhesion road surfaces. This reflects the differentiation while maintaining a reasonable range of protection.

[0049] It needs to be further explained that, Figure 3 This is a control flowchart for road condition recognition and mode switching in this invention. The system collects sensor data at a frequency of 50Hz. After filtering and feature extraction, it performs multi-sensor fusion decision-making using DS evidence theory. When the road condition confidence level is higher than 0.7, it is compared with the current mode; if they are inconsistent, a safety constraint check (vehicle speed, motor temperature, SOC, etc.) is performed; after passing the check, the system gradually transitions to the target mode parameters within 0.5 to 2 seconds via a smoothing transition unit, and updates the current mode, repeating the process cyclically.

[0050] It needs to be further explained that, Figure 4 This is a flowchart illustrating the SOC adaptive logic in the long downhill mode of this invention. When the system detects a long downhill road condition, it first reads the current battery SOC output by the BMS. If the SOC is ≥ 90%, an alarm is issued and switching to long downhill mode is prohibited, the energy recovery intensity is forced to 0, and the driver is prompted to use the mechanical brakes, while continuously monitoring the SOC; when the SOC drops below 90%, it automatically switches to long downhill mode and resumes strong energy recovery. If the initial SOC is below 90%, it directly switches to long downhill mode.

[0051] The system of this invention effectively solves the problems of poor adaptability of fixed parameters, lag in manual switching, and insufficient recognition dimensions of a single sensor under complex mining road conditions. It realizes closed-loop control of the entire process from road condition perception, fusion decision, safety verification to smooth parameter transition, which significantly improves the power, safety and driving smoothness of electric mining trucks under harsh road conditions.

[0052] Example 3 To achieve the above invention, such as Figure 5 As shown, this embodiment also provides an adaptive driving mode switching control device 10 for electric mining trucks, which includes: The fusion identification module 100 is used to acquire multi-source sensor signals and battery state of charge (SOC) of electric mining trucks. After filtering, it extracts features such as slope, flatness, road surface type and adhesion coefficient. It inputs each feature into the DS evidence theory for fusion decision and outputs the current road condition category and confidence level.

[0053] The safety decision module 200 is used to perform a safety constraint check based on the battery state of charge (SOC) when the confidence level exceeds the threshold and the road condition category is inconsistent with the current mode: if the road condition is a long downhill and the SOC is greater than or equal to the high charge threshold, then the long downhill mode is prohibited, energy recovery is set to zero and an alarm is triggered. The prohibition is lifted and the mode is activated after the SOC falls below the threshold. If the check passes, the target control parameter group corresponding to the road condition is called.

[0054] The smooth execution module 300 is used to configure the torque response slope, maximum vehicle speed limit and energy recovery intensity in the target control parameter group with transition times, and then smoothly transition them to the target values ​​through a linear ramp function before outputting them to the motor controller, vehicle controller and battery management system for execution.

[0055] The loop control module 400 is used to execute the above steps in a preset cycle.

[0056] The device in this embodiment of the invention realizes closed-loop control of the entire process from road condition perception, fusion decision, safety verification to smooth parameter transition through a modular architecture. It effectively solves the problems of poor adaptability of fixed parameters and switching shock under complex mining road conditions, and significantly improves the adaptive capability and driving safety of electric mining trucks under harsh road conditions.

[0057] Example 4 To implement the methods of the above embodiments, the present invention also provides a computer device, such as... Figure 6 As shown, the computer device 600 includes a memory 601 and a processor 602; wherein, the processor 602 reads the executable program code stored in the memory 601 to run a program corresponding to the executable program code, so as to implement the various steps of the above-described electric mining truck adaptive driving mode switching control method.

[0058] Example 5 To implement the above embodiments, this application also proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements an adaptive driving mode switching control method for electric mining trucks as described in the foregoing embodiments.

[0059] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0060] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for adaptive driving mode switching control of an electric mining truck, characterized in that, include: The system acquires multi-source sensor signals and battery state of charge (SOC) of electric mining trucks. After filtering, it extracts features such as slope, smoothness, road surface type and adhesion coefficient. The features are then input into the DS evidence theory for fusion and decision-making, and the current road condition category and confidence level are output. When the confidence level exceeds the threshold and the road condition category is inconsistent with the current mode, a safety constraint check is performed based on the battery state of charge (SOC): if the road condition is a long downhill and the SOC is greater than or equal to the high charge threshold, then switching to the long downhill mode is prohibited, energy recovery is reset to zero and an alarm is triggered. Once the SOC falls below the threshold, the prohibition is lifted and the mode can be switched to. If the check passes, the target control parameter group corresponding to the road condition is called. The torque response slope, maximum vehicle speed limit, and energy recovery intensity in the target control parameter group are configured with transition times, and after being smoothly transitioned to the target value through a linear ramp function, they are output to the motor controller, vehicle controller, and battery management system for execution. The above steps are executed repeatedly at a preset cycle.

2. The method as described in claim 1, characterized in that, Acquire multi-source sensor signals from electric mining trucks, including: At a sampling frequency of no less than 50Hz, the longitudinal acceleration, pitch angle and vertical acceleration signals output by the six-axis IMU, the road surface depth image and visible light image signal output by the binocular stereo camera, the wheel speed signals of each drive wheel output by the wheel slip rate sensor, and the road surface temperature and humidity signals output by the road surface temperature and humidity sensor are collected respectively. The system receives the battery state of charge (SOC) output by the battery management system (BMS) via the vehicle's CAN bus, as well as the current vehicle speed and motor temperature fed back by the motor controller (MCU), which together with these signals constitute a multi-source sensing signal.

3. The method as described in claim 1, characterized in that, After filtering, features such as slope, smoothness, road surface type, and adhesion coefficient are extracted, including: The multi-source sensor signals are input into the signal acquisition and conditioning unit, and after Kalman filtering and noise reduction, a pre-processed multi-dimensional sensor dataset is generated. The real-time slope value is calculated based on the pitch angle in the multidimensional sensor dataset, and will remain for more than 3 seconds. The slope value is used as a candidate for slope road condition and output as a slope feature. The variance is calculated based on the vertical acceleration in the multidimensional sensor dataset, and the variance value is greater than... It is used as a candidate for road surface roughness and output as a feature of road surface smoothness; The depth images and visible light images in the multidimensional sensing dataset are input into a lightweight convolutional neural network. After classification and reasoning, the network outputs visual confidence vectors for four types of road surfaces: water accumulation, mud, snow and ice, and normal road surfaces, which are used as road surface type features. The slip ratio is calculated based on the wheel speed of each drive wheel in the multidimensional sensor dataset, and the adhesion coefficient feature is derived by combining the road surface temperature and humidity.

4. The method as described in claim 1, characterized in that, The various features are input into the DS evidence theory for fusion and decision-making, outputting the current road condition category and confidence level, including: Slope characteristics, road surface smoothness characteristics, road surface type characteristics, and adhesion coefficient characteristics are each constructed as an independent evidence body, and a corresponding basic confidence assignment function is assigned to each evidence body. The basic confidence assignment function of each piece of evidence is input into the DS evidence theory fusion center, and the Dempster synthesis rule is used for fusion to output the comprehensive confidence of each candidate road condition category. Select the one with the highest overall confidence level that exceeds the preset threshold. The candidate road condition categories are output as the current road condition category, and the corresponding comprehensive confidence level is output as the confidence level of the corresponding decision result.

5. The method as described in claim 1, characterized in that, Perform safety constraint checks based on the battery's state of charge (SOC), including: When the current road condition is a long downhill slope, the current SOC value output in real time by the Battery Management System (BMS) is read, and the current SOC value is compared with a preset high charge threshold. Compare; If the current SOC value is greater than or equal to Then the mode decision unit generates a prohibition command to prohibit entering the long downhill mode and a voice alarm command that the battery is too high and the electric braking will weaken. At the same time, it forces the target energy recovery intensity to be set to zero and outputs it to the battery management system (BMS). The current SOC value is continuously read at a preset monitoring period. When the current SOC value falls below a certain level... When the time comes, the mode decision unit automatically cancels the prohibition command, restores the strong energy recovery setting, determines the current road condition category as long downhill mode, and calls the corresponding target control parameter group.

6. The method as described in claim 1, characterized in that, The target value is smoothly transitioned to a linear ramp function, including: Read the current control parameter set in the current active mode, which includes the current torque response slope, the current maximum vehicle speed limit, and the current energy recovery intensity. The target torque response slope, target maximum vehicle speed limit, and target energy recovery intensity are read from the target control parameter set, and a preset transition time is configured for each control parameter. Preset transition time Set according to parameter type Within seconds; The current value, the target value, and their respective preset transition times are set. Substitute into the linear ramp function: in The current value, For the target value, The preset transition time, The time is counted from the start of the switchover and ; The control parameters are calculated in real time at the current moment according to a preset calculation cycle. Value, when When all control parameters reach their corresponding target values, a smooth transition is achieved.

7. The method as described in claim 1, characterized in that, In the DS evidence theory fusion judgment, when the current road condition category is frost-prone road, a multi-feature joint judgment method is adopted, including: Extract the rut depth and lateral unevenness values ​​of the road surface ahead from the depth image output by the binocular stereo camera; Extract the peak-to-peak value of the fluctuation from the vertical acceleration signal output by the six-axis IMU; When the rut depth value is greater than Lateral unevenness value greater than And the peak-to-peak value of the vertical acceleration fluctuation is greater than At that time, the above three judgment conditions are input into the DS Evidence Theory Fusion Center as independent evidence of the frost-prone road. The DS Evidence Theory Fusion Center performs Dempster synthesis rule fusion on the three evidence bodies and outputs a comprehensive confidence score for the sluice gate category. When the comprehensive confidence score exceeds a preset threshold... At that time, the current road condition category output is "flooded road".

8. An adaptive driving mode switching control device for electric mining trucks, characterized in that, include: The fusion recognition module is used to acquire multi-source sensor signals and battery state of charge (SOC) of electric mining trucks. After filtering, it extracts features such as slope, flatness, road surface type and adhesion coefficient. The features are then input into the DS evidence theory for fusion judgment, and the current road condition category and confidence level are output. The safety decision module is used to perform a safety constraint check based on the battery state of charge (SOC) when the confidence level exceeds the threshold and the road condition category is inconsistent with the current mode: if the road condition is a long downhill and the SOC is greater than or equal to the high charge threshold, then the long downhill mode is prohibited, energy recovery is set to zero and an alarm is triggered. The prohibition is lifted and the mode is enabled after the SOC falls below the threshold. If the check passes, the target control parameter group corresponding to the road condition is called. The smooth execution module is used to configure the torque response slope, maximum vehicle speed limit and energy recovery intensity in the target control parameter group with transition time, and then smoothly transition them to the target value through a linear ramp function, and output them to the motor controller, vehicle controller and battery management system for execution respectively. The loop control module is used to execute the above steps in a preset cycle.

9. An electronic device, characterized in that, include: processor; The memory stores executable instructions; when the processor executes the instructions, it implements the adaptive driving mode switching control method for electric mining trucks as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The device stores a computer program, which, when executed by a processor, implements an adaptive driving mode switching control method for electric mining trucks as described in any one of claims 1-7.

Images