Adaptive orbital robot motion control method and device

Abstract

This application provides an adaptive motion control method and apparatus for a tracked robot. The method includes: using a multi-sensor array mounted on the robot body to collect RFID absolute position, encoder relative position, IMU attitude data, and visual environment images in real time to form multi-source perception data; performing spatiotemporal alignment and fusion processing on the multi-source information through a data fusion algorithm to generate the robot's current operating state information; and adaptively generating motion control commands based on the operating state information, combined with task objectives and safety constraints: dynamically adjusting speed and torque output to maintain uniform speed operation when the slope changes; automatically and smoothly decelerating when curves are encountered; performing deceleration, stopping, or obstacle avoidance operations when obstacles are detected; and performing braking operations when slipping. This application can improve the accuracy and efficiency of tracked robot control.

Description

Technical Field

[0001] This application relates to the field of data processing, specifically to an adaptive orbital robot motion control method and device. Background Technology

[0002] Adaptive orbit robot motion control systems are mainly used in industrial automation, logistics warehousing, intelligent inspection and other fields, aiming to solve the outstanding problems of traditional orbit robots in terms of positioning accuracy, anti-interference ability, environmental adaptability and intelligence level.

[0003] Traditional track-based robots rely heavily on encoders or simple sensors for positioning, making them susceptible to slippage, wear, and other factors that accumulate errors, failing to meet the demands of high-precision operations. In terms of control, traditional PID controllers have limited parameter adjustment capabilities under complex conditions such as vibration, load variations, or uneven tracks, easily leading to oscillations or loss of control, resulting in insufficient system robustness. Furthermore, fixed-parameter control strategies struggle to adapt to dynamic environments such as slope changes and wind load interference, causing robot performance degradation or instability. Simultaneously, existing systems generally lack autonomous decision-making capabilities, unable to adjust motion strategies in real time according to environmental changes; for example, they may simply lock up when encountering obstacles, lacking flexible obstacle avoidance and speed adjustment mechanisms, resulting in a low level of intelligence.

[0004] In conclusion, there is an urgent need for an adaptive motion control method for tracked robots that can improve the accuracy and efficiency of tracked robot control. Summary of the Invention

[0005] To address the problems in the prior art, this application provides an adaptive orbital robot motion control method and apparatus, which can improve the accuracy and efficiency of orbital robot control.

[0006] To solve at least one of the above problems, this application provides the following technical solution: In a first aspect, this application provides an adaptive orbital robot motion control method, comprising: A multi-sensor array is deployed on the body of the track robot. Multi-source perception data during the operation of the track robot is collected in real time based on the multi-sensor array. The multi-source perception data is spatiotemporally synchronized to determine the corresponding spatiotemporally aligned perception data. The spatiotemporal alignment perception data is evaluated in real time to determine the confidence weight of each perception data. The confidence weight is fused with the spatiotemporal alignment perception data according to a preset extended Kalman filter to determine the corresponding robot fusion state vector. Based on the robot fusion state vector and the image data collected by the vision sensor, the corresponding robot current running state is determined, including the robot's current track slope state, curve state, obstacle state, and slippage / derailment state. Based on the operating status, combined with the preset task objectives and safety constraints, corresponding adaptive motion control commands are determined and sent to the robot execution layer to achieve robot motion control. The generation process of the adaptive motion control commands includes: when the robot is detected to be in a slope change section, adjusting the speed and torque output according to the slope; when the robot is detected to be in a curve, generating smooth acceleration and deceleration commands; when an obstacle is detected on the track ahead, generating deceleration or stopping commands; and when a risk of slippage and derailment is detected, generating braking commands.

[0007] Furthermore, the step of collecting multi-source perception data in real time during the operation of the orbital robot based on the multi-sensor array includes: The absolute position data of key nodes on the track is collected using RFID readers. The robot's relative position and speed data are collected based on the built-in photoelectric encoder in the motor; The robot's attitude data, including pitch and roll angles, are collected using a six-axis inertial measurement unit. Image data of the environment in front of the track is collected using a visual sensor.

[0008] Furthermore, the real-time quality assessment includes: The confidence level of the data collected by the RFID reader is adjusted based on the track node tags, including assigning the highest confidence level to the absolute position data when it is determined that the RFID reader has read a tag. The confidence level of the data collected by the photoelectric encoder is adjusted based on the robot's slipping state, including assigning a first confidence level when there is no slipping and assigning a second confidence level when slipping is detected, wherein the second confidence level is lower than the first confidence level. The confidence level of the data collected by the six-axis inertial measurement unit is adjusted based on the robot's motion state, including assigning a third confidence level when the robot is stationary or moving at a constant speed, and assigning a fourth confidence level when the robot is vibrating violently. The fourth confidence level is lower than the third confidence level. The confidence level of the data collected by the visual sensor is adjusted based on the lighting conditions and image clarity. This includes assigning a fifth confidence level when the lighting is sufficient and the image is clear, and assigning a sixth confidence level when the lighting is insufficient or the image is blurry. The sixth confidence level is lower than the fifth confidence level.

[0009] Furthermore, the step of adjusting the speed and torque output according to the slope magnitude when the robot is detected to be in a slope change section includes: When the robot enters the ramp and recognizes the ramp label, it determines the corresponding entry reference torque based on the preset ramp slope value, preset ramp speed curve and current load current value, so that the robot can smoothly enter the ramp at the preset target speed. When the robot enters the ramp and goes uphill, the pitch angle data of the robot body is obtained in real time from the six-axis inertial measurement unit as the current slope value. The current slope value, real-time speed value and load current value are input to the preset fuzzy PID controller. The PID parameters are self-tuned online through fuzzy rules to determine the corresponding target torque so that the robot can pass the ramp at a constant speed according to the target torque. When the robot enters the ramp and goes downhill, the corresponding braking force compensation value is determined based on the slope value, real-time speed value and load current value at the time of descent, and the braking force compensation value is added to the target torque to prevent overspeed.

[0010] Furthermore, the step of generating smooth acceleration / deceleration commands when the robot is detected to be on a curve includes: When the robot is detected to be at the entrance of a curve, the radius of curvature and length of the current curve are determined based on the robot's absolute position data and the preset track map information. The corresponding curve speed limit value is determined based on the radius of curvature. A smooth deceleration command is triggered before entering the curve based on the curve speed limit value, so that the robot enters the curve at a speed lower than that of the straight section. When the robot is detected to be on a curve, the roll angle data fed back by the six-axis inertial measurement unit is acquired in real time. When the roll angle data exceeds a preset threshold, a speed suppression command is generated to prevent overturning. When the robot is detected to be at the exit of a curve, a smooth acceleration command is generated to restore the cruising speed to that of the straight section.

[0011] Furthermore, the step of generating a deceleration or stopping command when an obstacle is detected on the track ahead includes: When an obstacle is detected on the track ahead, the image data collected by the vision sensor is used to infer the corresponding obstacle type, obstacle distance, and the area of ​​the track occupied by the obstacle through a locally deployed lightweight target detection model. Based on obstacle type and obstacle distance, generate tiered response commands, including: When the distance to the obstacle exceeds a first preset threshold, a deceleration command is generated; When the distance to the obstacle is less than or equal to the first preset threshold and greater than the second preset threshold, a parking instruction is generated and reported to the dispatch center. When the distance to the obstacle is less than or equal to the second preset threshold, an emergency braking command is generated, triggering a hard emergency stop. Based on the information about the area occupied by the obstacle on the track, determine whether it is possible to detour. If it is possible to detour, then determine the corresponding low-speed detour operation instruction according to the detour instruction issued by the dispatch center.

[0012] Furthermore, the step of generating a braking command when a risk of slippage and derailment is detected includes: The drive wheel speed fed back by the photoelectric encoder is compared with the robot body acceleration fed back by the six-axis inertial measurement unit. When the deviation between the rate of change of speed and the rate of change of acceleration exceeds a preset threshold, it is determined that there is a risk of slippage. In the state of slippage risk, a torque limiting command is generated to reduce the output torque of the drive wheel and an alarm message is generated and reported to the dispatch center. The roll angle fed back by the six-axis inertial measurement unit is compared with the preset track tilt angle model. When the roll angle deviates from the preset track attitude by more than a threshold, it is determined that there is a risk of derailment. In the case of derailment risk, an emergency stop command is immediately generated and the electromagnetic power failure brake is triggered to lock the drive wheels.

[0013] Secondly, this application provides an adaptive orbital robot motion control device, comprising: The multi-source data spatiotemporal alignment module is used to deploy a multi-sensor array on the body of the track robot, collect multi-source perception data in real time during the operation of the track robot according to the multi-sensor array, perform spatiotemporal synchronization on the multi-source perception data, and determine the corresponding spatiotemporal aligned perception data. The track robot state judgment module is used to perform real-time quality assessment on the spatiotemporal alignment perception data, determine the confidence weight of each perception data, fuse the confidence weight with the spatiotemporal alignment perception data according to a preset extended Kalman filter, determine the corresponding robot fusion state vector, and determine the corresponding robot current running state according to the robot fusion state vector and the image data collected by the visual sensor, including the current track slope state, curve state, obstacle state and slippage / derailment state of the robot. The adaptive motion control command generation module is used to determine the corresponding adaptive motion control command based on the operating state, combined with preset task objectives and safety constraints, and send the motion control command to the robot execution layer to realize robot motion control. The generation process of the adaptive motion control command includes: when the robot is detected to be in a slope change section, adjusting the speed and torque output according to the slope; when the robot is detected to be in a curve, generating a smooth acceleration or deceleration command; when an obstacle is detected on the track ahead, generating a deceleration or stopping command; and when a risk of slippage and derailment is detected, generating a braking command.

[0014] Thirdly, this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the adaptive orbital robot motion control method.

[0015] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the adaptive orbital robot motion control method.

[0016] Fifthly, this application provides a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the adaptive orbital robot motion control method.

[0017] As can be seen from the above technical solution, this application provides an adaptive orbital robot motion control method and device. The robot body is equipped with a multi-sensor array to collect RFID absolute position, encoder relative position, IMU attitude data, and visual environment images in real time, forming multi-source perception data. A data fusion algorithm is used to perform spatiotemporal alignment and fusion processing on the multi-source information to generate the robot's current operating state information. Based on the operating state information, combined with the task objectives and safety constraints, motion control commands are adaptively generated: dynamically adjusting speed and torque output to maintain uniform speed operation when the slope changes; automatically and smoothly decelerating when curves are encountered; performing deceleration, stopping, or obstacle avoidance operations when obstacles are detected; and performing braking operations when slipping. This improves the accuracy and efficiency of orbital robot control. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is one of the flowcharts illustrating the adaptive orbital robot motion control method in the embodiments of this application; Figure 2 This is a structural diagram of the adaptive orbital robot motion control device in the embodiments of this application; Figure 3 This is a schematic diagram of the structure of the electronic device in the embodiments of this application.

[0020] Figure label: Electronic device 9600, central processing unit 9100, memory 9140, communication module 9110, input unit 9120, audio processor 9130, display 9160, power supply 9170, buffer memory 9141, application / function storage unit 9142, data storage unit 9143, driver storage unit 9144, antenna 9111, speaker 9131, microphone 9132. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0022] The acquisition, storage, use, and processing of data in this application comply with relevant laws and regulations.

[0023] Considering the current challenges faced by track-based robots, such as large cumulative positioning errors and limited parameter tuning capabilities of PID controllers based on fixed parameters, this application provides an adaptive motion control method and device for track-based robots. By equipping the robot with a multi-sensor array, it collects real-time data on RFID absolute position, encoder relative position, IMU attitude, and visual environment images, forming multi-source perception data. A data fusion algorithm is then used to perform spatiotemporal alignment and fusion processing on this multi-source information to generate the robot's current operating state information. Based on this operating state information, combined with task objectives and safety constraints, adaptive motion control commands are generated: dynamically adjusting speed and torque output to maintain uniform speed during slope changes; automatically and smoothly decelerating on curves; performing deceleration, stopping, or obstacle avoidance operations when obstacles are detected; and braking operations when slippage occurs. This improves the accuracy and efficiency of track-based robot control.

[0024] To improve the accuracy and efficiency of orbital robot control, this application provides an embodiment of an adaptive orbital robot motion control method, see [link to embodiment]. Figure 1 The adaptive orbital robot motion control method specifically includes the following: Step S101: Deploy a multi-sensor array on the track robot body, collect multi-source perception data in real time during the operation of the track robot according to the multi-sensor array, perform spatiotemporal synchronization on the multi-source perception data, and determine the corresponding spatiotemporal aligned perception data. Optionally, in this embodiment, the entire embodiment is applied to a track robot, and preferably, it is applied to an aerial track robot.

[0025] The communication architecture in this embodiment is as follows: End-side (robot): Built-in embedded motherboard, running lightweight control algorithms and AI vision models; Side-side (field-end server): Processes massive amounts of data (video streams, point cloud data) transmitted back by the robot and runs complex weight estimation and inventory algorithms; Cloud (Management Platform): Responsible for global task scheduling, data storage, and visualization.

[0026] Communication protocol: The robot maintains a stable connection with the cloud via industrial Wi-Fi and uses the MQTT protocol for command and data exchange.

[0027] To accurately perceive the position and self-state information of the track robot, avoid inaccurate positioning by a single sensor and external interference, and better control the speed and state of the track robot, we deployed a multi-sensor array composed of different types of sensors on the track robot body. The array includes an RFID reader for acquiring absolute position, an optical encoder for measuring relative displacement and velocity, a six-axis inertial measurement unit (IMU) for sensing the robot's own attitude, and a vision sensor for observing the external environment (such as obstacles in front of the track).

[0028] When the robot begins to run on the track, all the aforementioned sensors are activated, enabling them to synchronously and continuously collect multi-source sensor data during the robot's operation. The multi-source sensor data are physically independent; for example, RFID provides discrete location tag information, the encoder provides continuous odometer pulse data, the IMU provides three-axis acceleration and angular velocity data, and the vision sensor provides images or video streams. Because each sensor has a different sampling frequency, activation time, and data transmission delay, this raw data is misaligned in the time dimension and inconsistent in the spatial reference system (e.g., visual data is based on an image coordinate system, while IMU data is based on the robot's body coordinate system).

[0029] To address the aforementioned misalignment issue, spatiotemporal alignment of multi-source data is performed.

[0030] In the time dimension, each frame of data from different sensors is stamped with a uniform, high-precision timestamp (usually based on the system clock), and data streams of different frequencies are mapped onto the same time axis through interpolation or alignment algorithms, ensuring that the data used for subsequent calculations reflects the robot's state at the same moment. In the spatial dimension, using pre-calibrated sensor extrinsic parameters (i.e., the relative positions and attitude relationships between sensors), all perceived data is transformed into the same unified reference coordinate system (e.g., the robot's body coordinate system or the world coordinate system).

[0031] Through the above spatiotemporal synchronization processing, the original multi-source data is integrated into spatiotemporally aligned sensing data, providing a data foundation for subsequent data fusion, state recognition, and motion decision-making.

[0032] Step S102: Perform real-time quality assessment on the spatiotemporal alignment perception data, determine the confidence weight of each perception data, fuse the confidence weight with the spatiotemporal alignment perception data according to the preset extended Kalman filter, determine the corresponding robot fusion state vector, and determine the corresponding robot current running state according to the robot fusion state vector and the image data collected by the visual sensor, including the robot's current track slope state, curve state, obstacle state, and slippage / derailment state. Optionally, in this embodiment, real-time quality assessment is performed on the spatiotemporally aligned multi-source sensing data.

[0033] The data quality of each type of sensor is dynamically evaluated, with evaluation indicators including data validity, noise level, signal stability, and the presence of packet loss or abnormal jumps. Based on the evaluation results, the system assigns confidence weights to each sensor's data: sensors with higher data quality, lower noise, and more stable historical performance have higher confidence weights; conversely, when data quality is poor or anomalies occur, their weights are automatically reduced to minimize the negative impact of unreliable data on the fusion results.

[0034] Subsequently, the data from each sensor and their corresponding confidence weights are input into the Extended Kalman Filter (EKF). The EKF is a recursive filtering algorithm suitable for nonlinear systems, capable of optimal estimation of noisy observation data. Here, the filter predicts the robot's state based on its kinematic model, then corrects the prediction by incorporating the observation data with confidence weights, ultimately outputting a fused robot state vector. This state vector integrates multi-source information, making it more accurate, smoother, and more reliable than data from a single sensor.

[0035] Based on this, the system combines the robot's fused state vector with image data collected by the vision sensor to further identify and determine the robot's specific operating state. The two complement each other, jointly determining: track slope state (whether the robot is currently on an uphill or downhill section and the magnitude of the slope); curve state (whether it has entered a curve and the curvature of the curve); obstacle state (whether there are obstacles on the track ahead and their relative distance); and slippage / derailment state (by comparing the speed feedback from the encoder with the attitude changes feedback from the IMU to determine if there is a risk of wheel spin, slippage, or derailment).

[0036] By using quality assessment and confidence weighting, the anti-interference capability and robustness of data fusion are significantly improved; by using extended Kalman filtering, optimal fusion of multi-source data is achieved, eliminating measurement noise and cumulative errors from single sensors; the final output of operating status information is accurate, comprehensive, and continuous, providing reliable status input for subsequent adaptive motion decisions.

[0037] Step S103: Based on the operating state, combined with the preset task objective and safety constraints, determine the corresponding adaptive motion control command, and send the motion control command to the robot execution layer to realize robot motion control. The generation process of the adaptive motion control command includes: when the robot is detected to be in a slope change section, adjusting the speed and torque output according to the slope; when the robot is detected to be in a curve, generating a smooth acceleration and deceleration command; when an obstacle is detected on the track ahead, generating a deceleration or stopping command; when a risk of slippage and derailment is detected, generating a braking command.

[0038] Optionally, in this embodiment, the robot can automatically adjust its motion strategy according to adaptive motion control commands under different working conditions, thereby achieving smooth and safe track operation.

[0039] First, when the robot identifies itself as being on a slope using IMU attitude data, the decision unit dynamically adjusts the speed and torque output of the servo drive system based on the slope. Fuzzy PID control automatically increases torque to prevent speed loss when going uphill and increases braking force to prevent "rushing" when going downhill, thus ensuring the robot smoothly traverses the slope at a constant speed and avoiding "nodding" when climbing or "sprinting" when going downhill. Second, when the robot determines that it is entering a curve by fusing data from the encoder and IMU, the decision unit will generate smooth acceleration and deceleration commands, so that the robot automatically decelerates before entering the curve and automatically accelerates after exiting the curve, avoiding the risk of the robot becoming unstable or derailing due to centrifugal force generated by high-speed cornering.

[0040] Third, when the robot detects an obstacle on the track ahead using its visual sensors, the decision-making unit generates corresponding deceleration or stopping commands based on the distance and type of the obstacle. If the obstacle is a temporary, waitable target, the robot will stop at the designated point and wait; if it is a persistent obstacle, the robot will report to the dispatch center and perform a detour or reverse operation, rather than simply deadlocking.

[0041] Fourth, when the decision unit detects an abnormal speed feedback from the encoder but no corresponding change in the IMU attitude (such as the wheels spinning freely), it determines that there is a risk of slippage or derailment, immediately generates a braking command, triggers the electromagnetic de-energizer to lock the drive wheels, and prevents the robot from slipping out of control.

[0042] This example demonstrates how this embodiment obtains an accurate real-time state of the tracked robot based on multi-sensor data fusion, and performs adaptive motion control based on the real-time state.

[0043] As described above, the adaptive track robot motion control method provided in this application can collect RFID absolute position, encoder relative position, IMU attitude data, and visual environment images in real time through a multi-sensor array mounted on the robot body, forming multi-source perception data. A data fusion algorithm is then used to perform spatiotemporal alignment and fusion processing on the multi-source information to generate the robot's current operating status information. Based on this operating status information, combined with task objectives and safety constraints, motion control commands are adaptively generated: dynamically adjusting speed and torque output to maintain uniform speed operation when the slope changes; automatically and smoothly decelerating when curves are encountered; performing deceleration, stopping, or obstacle avoidance operations when obstacles are detected; and braking operations when slipping. This improves the accuracy and efficiency of track robot control.

[0044] In one embodiment of the adaptive orbital robot motion control method of this application, it may further include the following: Step S201: Collect absolute position data at key nodes on the track using the RFID reader; Step S202: Collect the robot's relative position and speed data based on the built-in photoelectric encoder in the motor; Step S203: Collect robot body attitude data, including pitch angle and roll angle, based on the six-axis inertial measurement unit; Step S204: Collect image data of the environment in front of the track using a vision sensor.

[0045] Optionally, in this embodiment, the absolute position data of key track nodes are obtained through an RFID reader.

[0046] At specific key nodes along the robot's track, such as the start and end points, intersections, ramp start and end points, curve entrances and exits, charging station locations, or observation points requiring precise positioning, RFID tags with unique identification codes are pre-embedded or affixed. The robot itself is equipped with an RFID reader. As the robot travels along the track and passes over these tags, the reader non-contactly reads the identification code and preset position information stored within the tag using radio frequency signals. The system then obtains the robot's absolute geographical location on the track at that moment. Because the robot accumulates errors when relying on encoders for relative positioning, the system can reset or calibrate the encoder's odometer data using the read absolute position information after each RFID tag pass, thus completely eliminating the accumulated positional deviation over long-term operation and ensuring that the robot can consistently and accurately know its "position on the track."

[0047] Optionally, in this embodiment, relative position and velocity data are acquired using a photoelectric encoder.

[0048] A high-precision photoelectric encoder is installed on the drive motor shaft or the axle of the walking wheel of the robot. When the motor rotates, the code disk inside the encoder, in conjunction with the light source and photosensitive element, converts the angular displacement of the shaft into a series of electrical pulse signals. The encoder outputs one pulse for each fixed small rotation. The robot control system counts the number of pulses to calculate the number of revolutions of the motor or wheel, and then, combined with the wheel circumference, calculates the relative distance traveled by the robot. Simultaneously, by measuring the pulse frequency per unit time, the robot's running speed can be calculated in real time. Between two RFID tags, the encoder provides precise continuous displacement information, filling the gap in absolute positioning and enabling continuous, high-resolution relative positioning and speed perception for the robot.

[0049] Optionally, in this embodiment, the robot's posture data is acquired using a six-axis inertial measurement unit.

[0050] A six-axis inertial measurement unit (IMU) is rigidly mounted inside the robot body. This unit integrates three orthogonal accelerometers and three orthogonal gyroscopes. The accelerometers measure the linear acceleration components of the robot body along each axis under the influence of gravity, and by solving these components, the robot's roll angle (left / right tilt angle) and pitch angle (forward / backward pitch angle) can be obtained. The gyroscopes measure the angular velocity of the robot's rotation around the three axes, which is used to assist in attitude calculation and suppress vibration noise, enabling the system to perceive the robot's attitude in space in real time.

[0051] When the robot is climbing or descending a slope, the pitch angle data clearly indicates the magnitude and direction of the slope; when the robot tilts due to uneven tracks or turning, the roll angle data reflects this in a timely manner. This attitude data is used for two purposes: firstly, for compensation by the motion controller (e.g., adjusting motor torque according to the slope), and secondly, for safety monitoring. When the attitude angle exceeds a safety threshold, it triggers a warning or emergency braking to prevent the robot from tipping over.

[0052] Optionally, in this embodiment, image data of the environment in front of the track is acquired using a visual sensor.

[0053] One or more vision sensors, such as monocular cameras, binocular cameras, or depth cameras, are mounted at the front end of the robot's path of travel (usually at the front of the robot). The sensor lenses face forward along the track, continuously capturing images or video streams along the track's direction of travel at a specific field of view. The image data is transmitted in real-time as digital signals to the robot's local embedded processing unit. Depending on the application requirements, the vision sensors can operate in visible light or infrared mode to adapt to different ambient lighting conditions.

[0054] Through step S204, this embodiment successfully acquired multi-source operational data of the track robot by equipping it with multiple sensors, laying the foundation for subsequent accurate analysis of the robot's state and implementation of motion control.

[0055] In one embodiment of the adaptive orbital robot motion control method of this application, it may further include the following: Step S301: Adjust the confidence level of the data collected by the RFID reader based on the track node tag, including assigning the highest confidence level to the absolute position data when it is determined that the RFID reader has read the tag; Step S302: Adjust the confidence level of the data collected by the photoelectric encoder based on the robot's slipping state, including assigning a first confidence level when there is no slipping and assigning a second confidence level when slipping is detected, wherein the second confidence level is lower than the first confidence level; Step S303: Adjust the confidence level of the data collected by the six-axis inertial measurement unit based on the robot's motion state, including assigning a third confidence level when stationary or moving at a constant speed, and assigning a fourth confidence level when vibrating violently, wherein the fourth confidence level is lower than the third confidence level; Step S304: Adjust the confidence level of the data collected by the visual sensor based on the lighting conditions and image clarity, including assigning a fifth confidence level when the lighting is sufficient and the image is clear, and assigning a sixth confidence level when the lighting is insufficient or the image is blurry, wherein the sixth confidence level is lower than the fifth confidence level.

[0056] Optionally, in this embodiment, RFID tags are pre-embedded or affixed at specific nodes of the track as the robot runs along it. When the RFID reader on the bottom of the robot passes over the tag, it triggers a reading operation to obtain a unique tag code, which corresponds to an absolute geographical location on the track.

[0057] The system assigns a confidence level to the RFID reading results. The moment the RFID reader successfully reads the tag is considered the exact moment the robot passes that track node. Since the RFID tag's position is precisely pre-calibrated, and the reading process is unaffected by common interference factors such as wheel slippage and sensor drift, the reliability of this absolute position information is extremely high. Therefore, the system assigns the highest confidence level to the absolute position data acquired at this moment. An absolutely reliable position reference benchmark is established in the data fusion system. This high-confidence data can be used to periodically zero or calibrate the mileage data accumulated by other sensors (such as photoelectric encoders), fundamentally eliminating accumulated positioning errors caused by long-term operation and ensuring the long-term stability of the robot's positioning system.

[0058] Optionally, in this embodiment, the photoelectric encoder calculates the distance and speed traveled by the robot by recording the number of rotations. Its basic assumption is that there is a fixed linear relationship between the number of wheel rotations and the actual distance traveled.

[0059] We comprehensively compare various data points to determine whether slippage has occurred. For example, we compare the speed calculated by the encoder with the integral acceleration measured by the IMU, or we compare the displacement calculated by the encoder with the position after RFID calibration. When the system determines that the robot is moving normally without slippage, the encoder data accurately reflects the actual motion, and a first confidence level (high confidence) is assigned. Conversely, when slippage is detected (such as wheels spinning freely on a wet track) or lockup (wheels not turning but the robot is still sliding), the encoder data deviates from the actual motion, and a second confidence level (low confidence) is assigned.

[0060] The dynamic identification of encoder data enhances its credibility, enabling upper-layer data fusion algorithms to automatically adjust the weight of the data source based on its confidence level.

[0061] Optionally, in this embodiment, the six-axis inertial measurement unit consists of a three-axis accelerometer and a three-axis gyroscope, used to measure the acceleration and angular velocity of the robot body. These raw measurements include gravitational components, motion acceleration, and various high-frequency vibration noises.

[0062] We comprehensively assess the robot's motion state using factors such as speed commands, encoder feedback, and acceleration variance. When the robot is determined to be stationary or in uniform linear motion, the acceleration values ​​measured by the IMU primarily originate from the gravitational component, resulting in relatively stable signals with low noise. The attitude and velocity information obtained through integration is relatively accurate, thus assigning a third confidence level (high confidence). When the robot is determined to be in a state of severe vibration (such as traveling at a track joint or experiencing external impact), the IMU output signal contains a large amount of high-frequency noise and transient interference. Direct use of this signal would lead to errors in attitude calculation, thus assigning a fourth confidence level (low confidence).

[0063] Identify the effective range of IMU data, and make full use of IMU data for attitude compensation and velocity-assisted calculation during stable operation; during vibration disturbance, actively reduce the fusion weight of IMU data and rely more on other sensors such as encoders or vision, thereby avoiding the negative impact of vibration noise on motion control decisions.

[0064] Optionally, in this embodiment, a vision sensor (such as a monocular or binocular camera) is responsible for acquiring image data of the area in front of and around the track for tasks such as obstacle detection and track recognition. Image quality directly determines the effectiveness of subsequent vision algorithms.

[0065] The system calculates multiple image quality parameters in real time, including: overall average brightness (to determine if the image is too dark or overexposed), brightness distribution uniformity (to determine if there are local shadows or strong lights), edge gradient energy (to determine if the image is clear and in focus), and motion blur (to determine if robot vibration causes ghosting). When the lighting is sufficient, the brightness is moderate, and the image edges are clear and without significant blur, it is judged as a high-quality image and assigned a fifth confidence level (high confidence). When the lighting is insufficient, the image is overexposed, severely out of focus, or motion blur causes loss of detail, it is assigned a sixth confidence level (low confidence).

[0066] It effectively prevents false detections and missed detections of visual algorithms under poor lighting or image degradation conditions, ensuring the stability and security of the entire perception system in various environments.

[0067] Through step S304, this embodiment realizes the dynamic assignment of confidence values ​​to different sensor data based on their current credibility, providing differentiated weighting basis for subsequent multi-source data fusion, thereby improving the accuracy and robustness of fusion positioning and perception results.

[0068] In one embodiment of the adaptive orbital robot motion control method of this application, it may further include the following: Step S401: When the robot enters the ramp and recognizes the ramp label, it determines the corresponding entry reference torque based on the preset ramp slope value, preset ramp speed curve and current load current value, so that the robot can smoothly enter the ramp at the preset target speed. Step S402: When the robot enters the ramp and goes uphill, the pitch angle data of the robot body is obtained in real time from the six-axis inertial measurement unit as the current slope value. The current slope value, real-time speed value and load current value are input to the preset fuzzy PID controller. The PID parameters are self-tuned online through fuzzy rules to determine the corresponding target torque so that the robot can pass the ramp at a constant speed according to the target torque. Step S403: When the robot enters the ramp and goes downhill, the corresponding braking force compensation value is determined based on the slope value, real-time speed value and load current value at the time of descent, and the braking force compensation value is added to the target torque to prevent overspeed.

[0069] Optionally, in this embodiment, a reference torque is set before entering the slope.

[0070] When the robot moves along the track to the preset position before the ramp entrance, the ramp tag installed on the track will be identified by the robot's RFID reader. At this time, the control system determines the reference torque required to enter the ramp based on three key parameters: first, the ramp's pre-stored slope value in the system map; second, the preset speed planning curve for this slope, i.e., the target speed to be maintained when entering the ramp; and third, the current load current value of the motor, which indirectly reflects the total load weight currently carried by the robot.

[0071] The three parameters mentioned above are input into the torque calculation model, which outputs a benchmark torque value that matches the current working conditions. The purpose of this torque value is to apply sufficient driving force in advance at the moment when the robot's front wheels just touch the ramp and the gravity component begins to increase, to prevent a sudden drop in speed or "nose-nodding" phenomenon caused by insufficient torque, so that the robot can smoothly and without impact enter the ramp at the preset target speed, and avoid the impact of sudden speed changes on the quality of sensor data.

[0072] Optionally, in this embodiment, after the robot fully enters the uphill section, the slope may not be a constant value, but rather exhibit slight fluctuations. Furthermore, based on the preferred embodiment, the real-time status of the aerial track robot also needs to consider wind load pressure. At this time, the control system reads the robot's real-time pitch angle from the six-axis inertial measurement unit at millisecond intervals, using this as the current true slope value. Simultaneously, it continuously collects the real-time speed value corresponding to the motor's rotational speed and the current load current value.

[0073] The three real-time parameters mentioned above are fed into a preset fuzzy PID controller. The core feature of this controller is that its PID parameters—proportional, integral, and derivative coefficients—can be automatically adjusted online according to changes in the input parameters. For example, when the slope suddenly becomes steeper, the controller automatically increases the proportional term to quickly increase the torque output; when the speed oscillates slightly, it automatically adjusts the derivative term to suppress the oscillation. Through this dynamic self-tuning, the controller continuously calculates and outputs the optimal target torque, enabling the robot to maintain a set uniform speed throughout the uphill climb, avoiding instability such as deceleration or sudden changes in speed during the climb.

[0074] Specifically, the implementation principle of the fuzzy PID controller is as follows: Traditional PID controllers have three parameters—proportional, integral, and derivative—that are set once before operation and remain unchanged during running. When the characteristics of the controlled object change, fixed-parameter PID controllers struggle to adapt, easily leading to problems such as overshoot, oscillation, or slow response.

[0075] The fuzzy PID controller introduces a fuzzy logic inference module into the traditional PID controller. This module monitors the deviation and rate of change of the controlled variable in real time, transforming these precise values ​​into linguistic descriptions such as "large deviation," "moderate deviation," and "small deviation" through "fuzzification." Subsequently, fuzzy inference is performed based on a pre-set expert rule base; for example, "if the deviation is large and the rate of change of the deviation is large, then significantly increase the proportional coefficient." The inference result is then defuzzified to output precise PID parameter adjustment values. These three parameters are continuously updated online during the control process, ensuring the controller always adapts to the current state of the controlled object, achieving self-tuning control.

[0076] To illustrate with a specific example, suppose a track robot is running at a constant speed of 0.5 m / s on a straight track. At this time, the parameters of the fuzzy PID controller are set as proportional coefficient Kp=10, integral coefficient Ki=2, and derivative coefficient Kd=1.

[0077] When the robot's front wheels contact a 15° ramp, the gravitational component begins to generate resistance. The control system detects that the current speed has decreased from 0.5 m / s to 0.45 m / s, and calculates a speed deviation of -0.05 m / s; at the same time, it detects that the rate of change of deviation is negative, indicating that the speed is still decreasing.

[0078] After receiving these two inputs, the fuzzy controller performs fuzzification: the deviation is determined to be "negatively medium" and the deviation change rate is determined to be "negatively small". According to the preset rule "if the deviation is negatively medium and the deviation change rate is negatively small, then the proportional coefficient increment should be positively medium", the system adjusts Kp from 10 to 15 to quickly increase torque output to compensate for the speed drop; at the same time, according to another rule "if the deviation change rate is negatively small, then the differential coefficient increment should be positively small", Kd is adjusted from 1 to 1.3 to suppress speed oscillations that may be caused by sudden torque increases.

[0079] The adjusted parameters rapidly increased the motor's output torque, and the robot's speed gradually recovered to 0.49 m / s. At this point, the deviation became negative, and the controller readjusted the parameters, reducing Kp back to 12 to prevent torque overshoot. After approximately 2-3 consecutive self-tuning adjustments, the robot's speed stabilized at 0.5 m / s, maintaining a uniform uphill speed. Throughout the entire process, the fuzzy PID controller achieved smooth control without overshoot or oscillation.

[0080] Optionally, in this embodiment, when the robot enters a downhill section, the gravity component changes from resistance to power, posing a risk of overspeed. The control system first calculates a basic braking force compensation value based on the current slope value, real-time speed value, and load current value. This compensation value is superimposed on the target torque output by the fuzzy PID controller in the form of a negative torque, which is equivalent to increasing the braking effect of the motor.

[0081] Specifically, the steeper the slope or the heavier the load, the greater the calculated braking force compensation value, thus suppressing acceleration caused by gravity. Simultaneously, the control system continuously monitors the real-time speed; if the speed continues to rise, the compensation value is dynamically increased; if the speed begins to fall below the target value, the compensation value is appropriately decreased. This closed-loop adjustment mechanism ensures that the robot does not experience "overspeeding" during downhill descents, always keeping the speed within a safe and stable range.

[0082] Through step S403, this embodiment successfully performed adaptive slope control for the track robot, achieving smooth operation of the track robot under constantly changing slope conditions.

[0083] In one embodiment of the adaptive orbital robot motion control method of this application, it may further include the following: Step S501: When the robot is detected to be at the entrance of a curve, the radius of curvature and length of the current curve are determined based on the robot's absolute position data and the preset track map information. The corresponding curve speed limit value is determined based on the radius of curvature. A smooth deceleration command is triggered before entering the curve based on the curve speed limit value so that the robot enters the curve at a speed lower than that of the straight section. Step S502: When the robot is detected to be on a curve, the roll angle data fed back by the six-axis inertial measurement unit is acquired in real time. When the roll angle data exceeds a preset threshold, a speed suppression command is generated to prevent overturning. Step S503: When the robot is detected to be at the exit of a curve, a smooth acceleration command is generated to restore the cruising speed to the straight section.

[0084] Optionally, in this embodiment, before the robot reaches the entrance of a curve, the system first obtains the robot's current absolute position via an RFID reader and, combined with the track map information pre-stored in the local decision unit, determines the radius of curvature and length of the curve ahead. The smaller the radius of curvature, the sharper the curve, and the lower the required speed for the robot to pass through. The system automatically calculates the maximum safe speed allowed for the curve based on the radius of curvature, which serves as the speed limit. Before the robot enters the curve, the control system triggers a smooth deceleration command based on this speed limit, causing the drive motor to gradually reduce its output speed, thus allowing the robot to smoothly enter the curve at a speed significantly lower than that of the straight cruising section. The purpose is to avoid the centrifugal force generated by high-speed entry into the curve causing the robot to tilt or derail, while providing safe initial speed conditions for subsequent curve driving. The smooth deceleration relies on the speed closed-loop control of the servo drive system to ensure that the deceleration process is shock-free and oscillating.

[0085] Optionally, in this embodiment, when the robot has entered the curve, the system continuously receives roll angle data fed back by the six-axis inertial measurement unit. The roll angle reflects the degree of tilt of the robot body around the forward axis and is a key indicator for judging the safety of curve driving. During normal curve driving, the robot will produce a certain centripetal tilt, but when the roll angle exceeds the system's preset safety threshold, it indicates that the robot may face the risk of tipping over. The cause may be excessive speed, track deformation, or uneven load. At this time, the control system immediately generates a speed suppression command, forcibly further reducing the output speed of the drive motor, or even issuing a point braking signal to reduce centrifugal force and restore the robot to a stable posture. The purpose of this step is to intervene dynamically in real time to prevent the robot from tipping over due to loss of posture in the curve, while avoiding impact caused by emergency braking.

[0086] Optionally, in this embodiment, when the robot is about to exit the curve, the system determines the exit position of the curve based on absolute position information and accumulated odometer data. At this time, the control system generates a smooth acceleration command, gradually increasing the output torque of the drive motor, so that the robot can smoothly accelerate from the low speed state of driving on the curve until it returns to the preset cruising speed on the straight section. This acceleration process adopts a ramp-type or S-curve-type acceleration strategy to avoid the impact and current overshoot caused by rapid acceleration. Its purpose is to ensure that the robot can efficiently enter the straight high-speed operation state after exiting the curve, improve the overall inspection efficiency, and at the same time maintain the stability and comfort of operation, protecting the mechanical structure and sensor equipment.

[0087] Through step S503, this embodiment successfully achieves adaptive curve control for the track robot, preventing overturning and ensuring that the robot completes curve segment operation smoothly and efficiently.

[0088] In one embodiment of the adaptive orbital robot motion control method of this application, it may further include the following: Step S601: When an obstacle is detected on the track ahead, the image data collected by the visual sensor is used to infer the corresponding obstacle type, obstacle distance, and obstacle-occupied track area information through a locally deployed lightweight target detection model. Step S602: Generate graded response instructions based on obstacle type and obstacle distance, including: When the distance to the obstacle exceeds a first preset threshold, a deceleration command is generated; When the distance to the obstacle is less than or equal to the first preset threshold and greater than the second preset threshold, a parking instruction is generated and reported to the dispatch center. When the distance to the obstacle is less than or equal to the second preset threshold, an emergency braking command is generated, triggering a hard emergency stop. Step S603: Determine whether the obstacle can be bypassed based on the information about the area occupied by the obstacle on the track. If it can be bypassed, determine the corresponding low-speed bypass action instruction based on the bypass instruction issued by the dispatch center.

[0089] Optionally, in this embodiment, when the robot detects an obstacle on the track ahead, the system first initiates a local vision processing flow. The robot's onboard vision sensors (such as a depth camera or monocular camera) acquire image data of the area in front of the track in real time. This image data is fed into a lightweight target detection model running on the robot's embedded industrial computer. This model is pre-trained and can quickly identify common obstacle types (such as people, tools, etc.) under low computing power. The model outputs three key pieces of information: the type of obstacle, the actual distance between the obstacle and the robot, and the extent of the obstacle's occupation on the track's cross-section (e.g., whether it completely blocks the track or only partially occupies it). The core function of this step is to achieve fast, low-latency obstacle perception without relying on the cloud, providing structured input data for subsequent decision-making.

[0090] Optionally, in this embodiment, a graded response strategy is executed based on the obstacle distance output in the previous step, combined with preset two-level safety thresholds.

[0091] Level 1: When the distance to the obstacle is greater than the first preset threshold (e.g., 3 meters), the system determines that the current risk is low and only generates a deceleration command to reduce the robot's running speed, allowing more reaction time for possible subsequent stopping or detouring.

[0092] Level 2: When the distance to the obstacle is less than or equal to the first preset threshold but greater than the second preset threshold (e.g., 1 meter), the system determines the risk to be moderate, immediately generates a stop command, and makes the robot stop smoothly in front of the obstacle. At the same time, the obstacle information and the robot status are reported to the field or cloud dispatch center to request further instructions.

[0093] Level 3: When the distance to the obstacle is less than or equal to the second preset threshold, the system determines it to be an emergency danger state, skips the normal stopping procedure, directly generates an emergency braking command, triggers the robot's hard emergency stop device (such as an electromagnetic power-off brake), and achieves a forced stop within the shortest distance. The purpose of this hierarchical strategy is to balance safety and operational continuity, avoid unnecessary emergency stops that affect efficiency, and ensure absolute safety in truly dangerous situations.

[0094] Optionally, in this embodiment, after the system determines that it needs to stop or has already stopped, it further assesses the feasibility of detour based on the information about the area occupied by the obstacle on the track. If the obstacle only partially occupies the track (e.g., there is space to pass on one side), and the track structure and robot mechanical design allow for detour, the system waits for a detour instruction from the scheduling center. Upon receiving the instruction, the system generates specific low-speed detour instructions, including a series of actions such as controlling the robot to move at a very low speed in the passable direction, bypassing the obstacle, and returning to the center of the track. The purpose of this step is to maintain the continuity of the robot's task as much as possible under safe and controllable conditions, avoiding prolonged downtime or manual intervention due to small obstacles.

[0095] Through step S603, this embodiment successfully achieves adaptive obstacle control for the track robot, maximizing the continuity and autonomy of the robot's operation while ensuring safety.

[0096] In one embodiment of the adaptive orbital robot motion control method of this application, it may further include the following: Step S701: Compare the drive wheel rotation speed fed back by the photoelectric encoder with the robot body acceleration fed back by the six-axis inertial measurement unit. When the deviation between the rotation speed change rate and the acceleration change rate exceeds a preset threshold, it is determined that there is a risk of slippage. Under the risk of slippage, a torque limiting command is generated to reduce the output torque of the drive wheel and an alarm message is generated and reported to the dispatch center. Step S702: Compare the roll angle fed back by the six-axis inertial measurement unit with the preset track tilt model. When the roll angle deviates from the preset track attitude by more than the threshold, it is determined that there is a risk of derailment. In the state of derailment risk, an emergency stop command is immediately generated and the electromagnetic power failure brake is triggered to lock the drive wheels.

[0097] Optionally, in this embodiment, during robot operation, the photoelectric encoder continuously measures the actual rotational speed of the drive wheels and calculates their rate of change; simultaneously, the six-axis inertial measurement unit (IMU) acquires the robot's acceleration data in real time and calculates the rate of change of acceleration. The system performs time synchronization and comparative analysis on these two sets of data.

[0098] When the rate of change of the drive wheel's rotational speed is significantly higher than the rate of change of the robot's acceleration, and the deviation between the two exceeds the system's preset safety threshold, the robot is determined to be in a slipping state. This deviation means that the drive wheel is spinning freely, while the robot body is not obtaining a corresponding speed increase. Typical scenarios include slippery tracks, insufficient wheel-rail friction, or sudden load changes.

[0099] Upon detecting a risk of slippage, the system automatically generates a torque limiting command, reducing the torque output from the servo drive system to the drive wheels in real time to prevent further slippage or wheel-rail wear caused by continuous high torque output. Simultaneously, the system generates a slippage alarm and reports it to the dispatch center via the communication module for recording and handling by operators or higher-level systems.

[0100] Optionally, in this embodiment, during robot operation, the six-axis inertial measurement unit continuously monitors the roll angle of the robot body, that is, the tilt angle of the robot around the forward axis. The system compares the measured roll angle with a preset track tilt angle model in real time, which describes the theoretical range and variation law of the roll angle of the robot under normal track operation.

[0101] When the measured roll angle deviates from the preset track posture by more than the system-set safety threshold, the robot is deemed to be at risk of derailment. This deviation indicates that the robot's posture has deviated from the normal track constraints, which may be caused by track deformation, wheels leaving the track, or external impact.

[0102] Upon detecting a derailment risk, the system immediately generates an emergency stop command, cutting off the power output of the drive motor and simultaneously triggering the electromagnetic power-off brake. This brake automatically engages in the power-off state, locking the drive wheels and bringing the robot to a rapid stop, preventing further movement that could lead to complete derailment or a fall. By comparing the roll angle with the track model, the system achieves early identification and quantitative assessment of derailment risks. The emergency stop and brake locking provide a hardware-level rapid response mechanism, minimizing the probability and severity of derailment accidents.

[0103] Through step S702, this embodiment successfully achieves adaptive risk state control of the track robot, minimizing the risk of accidents.

[0104] To improve the accuracy and efficiency of track robot control, this application provides an embodiment of an adaptive track robot motion control device for implementing all or part of the aforementioned adaptive track robot motion control method. See [link to embodiment]. Figure 2 The adaptive orbit robot motion control device specifically includes the following components: The multi-source data spatiotemporal alignment module 10 is used to deploy a multi-sensor array on the body of the track robot, collect multi-source perception data in real time during the operation of the track robot according to the multi-sensor array, perform spatiotemporal synchronization on the multi-source perception data, and determine the corresponding spatiotemporal aligned perception data. The track robot state judgment module 20 is used to perform real-time quality assessment on the spatiotemporal alignment perception data, determine the confidence weight of each perception data, fuse the confidence weight with the spatiotemporal alignment perception data according to a preset extended Kalman filter, determine the corresponding robot fusion state vector, and determine the corresponding robot current running state according to the robot fusion state vector and the image data collected by the visual sensor, including the current track slope state, curve state, obstacle state and slippage / derailment state of the robot. The adaptive motion control command generation module 30 is used to determine the corresponding adaptive motion control command based on the operating state, combined with the preset task objective and safety constraints, and send the motion control command to the robot execution layer to realize robot motion control. The generation process of the adaptive motion control command includes: when the robot is detected to be in a slope change section, adjusting the speed and torque output according to the slope; when the robot is detected to be in a curve, generating a smooth acceleration and deceleration command; when an obstacle is detected on the track ahead, generating a deceleration or stopping command; and when a risk of slippage and derailment is detected, generating a braking command.

[0105] As described above, the adaptive track robot motion control device provided in this application embodiment can collect RFID absolute position, encoder relative position, IMU attitude data, and visual environment images in real time through a multi-sensor array mounted on the robot body, forming multi-source perception data. The multi-source information is then spatiotemporally aligned and fused using a data fusion algorithm to generate the robot's current operating status information. Based on this operating status information, combined with task objectives and safety constraints, motion control commands are adaptively generated: dynamically adjusting speed and torque output to maintain uniform speed operation when the slope changes; automatically and smoothly decelerating when curves are encountered; performing deceleration, stopping, or obstacle avoidance operations when obstacles are detected; and braking operations when slipping. This improves the accuracy and efficiency of track robot control.

[0106] From a hardware perspective, in order to improve the accuracy and efficiency of orbital robot control, this application provides an embodiment of an electronic device for implementing all or part of the adaptive orbital robot motion control method, wherein the electronic device specifically includes the following: The system comprises a processor, memory, a communications interface, and a bus; wherein the processor, memory, and communications interface communicate with each other via the bus; the communications interface is used to realize information transmission between the adaptive orbit robot motion control method and core business systems, user terminals, and related databases and other related devices; the logic controller can be a desktop computer, tablet computer, or mobile terminal, etc., and this embodiment is not limited to these. In this embodiment, the logic controller can be implemented with reference to the embodiments of the adaptive orbit robot motion control method in the present embodiment, and the contents of the embodiments of the adaptive orbit robot motion control method are incorporated herein, and repeated details will not be described again.

[0107] It is understood that the user terminal may include smartphones, tablet computers, network set-top boxes, portable computers, desktop computers, personal digital assistants (PDAs), in-vehicle devices, smart wearable devices, etc. Among these, the smart wearable devices may include smart glasses, smartwatches, smart bracelets, etc.

[0108] In practical applications, the adaptive orbital robot motion control method can be partially executed on the electronic device side as described above, or all operations can be completed in the client device. The choice can be made based on the processing power of the client device and the limitations of the user's usage scenario. This application does not impose any limitations on this. If all operations are completed in the client device, the client device may further include a processor.

[0109] The aforementioned client device may have a communication module (i.e., a communication unit) that can communicate with a remote server to achieve data transmission. The server may include a server on the task scheduling center side; in other implementation scenarios, it may also include a server on an intermediate platform, such as a server on a third-party server platform that has a communication link with the task scheduling center server. The server may include a single computer device, a server cluster consisting of multiple servers, or a distributed server structure.

[0110] Figure 3 This is a schematic block diagram illustrating the system configuration of the electronic device 9600 according to an embodiment of this application. Figure 3 As shown, the electronic device 9600 may include a central processing unit 9100 and a memory 9140; the memory 9140 is coupled to the central processing unit 9100. It is worth noting that... Figure 3 This is an example; other types of structures can also be used to supplement or replace this structure to achieve telecommunications functions or other functions.

[0111] In one embodiment, the adaptive orbital robot motion control method functionality can be integrated into a central processing unit 9100. The central processing unit 9100 can be configured to perform the following control: Step S101: Deploy a multi-sensor array on the track robot body, collect multi-source perception data in real time during the operation of the track robot according to the multi-sensor array, perform spatiotemporal synchronization on the multi-source perception data, and determine the corresponding spatiotemporal aligned perception data. Step S102: Perform real-time quality assessment on the spatiotemporal alignment perception data, determine the confidence weight of each perception data, fuse the confidence weight with the spatiotemporal alignment perception data according to the preset extended Kalman filter, determine the corresponding robot fusion state vector, and determine the corresponding robot current running state according to the robot fusion state vector and the image data collected by the visual sensor, including the robot's current track slope state, curve state, obstacle state, and slippage / derailment state. Step S103: Based on the operating state, combined with the preset task objective and safety constraints, determine the corresponding adaptive motion control command, and send the motion control command to the robot execution layer to realize robot motion control. The generation process of the adaptive motion control command includes: when the robot is detected to be in a slope change section, adjusting the speed and torque output according to the slope; when the robot is detected to be in a curve, generating a smooth acceleration and deceleration command; when an obstacle is detected on the track ahead, generating a deceleration or stopping command; when a risk of slippage and derailment is detected, generating a braking command.

[0112] As described above, the electronic device provided in this application embodiment uses a multi-sensor array mounted on the robot body to collect RFID absolute position, encoder relative position, IMU attitude data, and visual environment images in real time, forming multi-source perception data. A data fusion algorithm is then used to perform spatiotemporal alignment and fusion processing on the multi-source information to generate the robot's current operating status information. Based on this operating status information, combined with task objectives and safety constraints, motion control commands are adaptively generated: dynamically adjusting speed and torque output to maintain uniform speed operation when the slope changes; automatically and smoothly decelerating when curves are encountered; performing deceleration, stopping, or obstacle avoidance operations when obstacles are detected; and braking operations when slipping. This improves the accuracy and efficiency of the track robot control.

[0113] In another embodiment, the adaptive orbit robot motion control method can be configured separately from the central processing unit 9100. For example, the adaptive orbit robot motion control method can be configured as a chip connected to the central processing unit 9100, and the function of the adaptive orbit robot motion control method can be realized through the control of the central processing unit.

[0114] like Figure 3 As shown, the electronic device 9600 may further include: a communication module 9110, an input unit 9120, an audio processor 9130, a display 9160, and a power supply 9170. It is worth noting that the electronic device 9600 does not necessarily need to include these components. Figure 3 All components shown; in addition, the electronic device 9600 may also include Figure 3 For components not shown, please refer to existing technology.

[0115] like Figure 3 As shown, the central processing unit 9100, sometimes also referred to as a controller or operating control, may include a microprocessor or other processor device and / or logic device, which receives inputs and controls the operation of various components of the electronic device 9600.

[0116] The memory 9140 may be, for example, one or more of a cache, flash memory, hard drive, removable media, volatile memory, non-volatile memory, or other suitable devices. It may store the aforementioned failure-related information, and also store a program for executing that information. The central processing unit 9100 may execute the program stored in the memory 9140 to perform information storage or processing, etc.

[0117] Input unit 9120 provides input to central processing unit 9100. Input unit 9120 may be, for example, a keypad or touch input device. Power supply 9170 provides power to electronic device 9600. Display 9160 displays images and text. Display may be, for example, an LCD display, but is not limited thereto.

[0118] The memory 9140 can be a solid-state memory, such as a read-only memory (ROM), random access memory (RAM), a SIM card, etc. It can also be a memory that retains information even when power is off, can be selectively erased, and contains more data; examples of this type of memory are sometimes referred to as EPROMs. The memory 9140 can also be some other type of device. The memory 9140 includes a buffer memory 9141 (sometimes referred to as a buffer). The memory 9140 may include an application / function storage unit 9142 for storing application programs and function programs or processes for executing the operation of the electronic device 9600 via the central processing unit 9100.

[0119] The memory 9140 may also include a data storage unit 9143 for storing data, such as contacts, digital data, pictures, sounds, and / or any other data used by the electronic device. The driver storage unit 9144 of the memory 9140 may include various drivers for the electronic device for communication functions and / or for performing other functions of the electronic device (such as messaging applications, address book applications, etc.).

[0120] The communication module 9110 is a transmitter / receiver that sends and receives signals via the antenna 9111. The communication module 9110 is coupled to the central processing unit 9100 to provide input signals and receive output signals, which is the same as in a conventional mobile communication terminal.

[0121] Based on different communication technologies, multiple communication modules 9110 can be configured in the same electronic device, such as cellular network modules, Bluetooth modules, and / or wireless LAN modules. The communication module 9110 is also coupled to a speaker 9131 and a microphone 9132 via an audio processor 9130 to provide audio output via the speaker 9131 and receive audio input from the microphone 9132, thereby realizing typical telecommunications functions. The audio processor 9130 may include any suitable buffer, decoder, amplifier, etc. Furthermore, the audio processor 9130 is also coupled to a central processing unit 9100, enabling on-device recording via the microphone 9132 and on-device playback of stored sound via the speaker 9131.

[0122] Embodiments of this application also provide a computer-readable storage medium capable of implementing all steps of the adaptive orbital robot motion control method with a server or client as the execution subject in the above embodiments. The computer-readable storage medium stores a computer program that, when executed by a processor, implements all steps of the adaptive orbital robot motion control method with a server or client as the execution subject in the above embodiments. For example, when the processor executes the computer program, it implements the following steps: Step S101: Deploy a multi-sensor array on the track robot body, collect multi-source perception data in real time during the operation of the track robot according to the multi-sensor array, perform spatiotemporal synchronization on the multi-source perception data, and determine the corresponding spatiotemporal aligned perception data. Step S102: Perform real-time quality assessment on the spatiotemporal alignment perception data, determine the confidence weight of each perception data, fuse the confidence weight with the spatiotemporal alignment perception data according to the preset extended Kalman filter, determine the corresponding robot fusion state vector, and determine the corresponding robot current running state according to the robot fusion state vector and the image data collected by the visual sensor, including the robot's current track slope state, curve state, obstacle state, and slippage / derailment state. Step S103: Based on the operating state, combined with the preset task objective and safety constraints, determine the corresponding adaptive motion control command, and send the motion control command to the robot execution layer to realize robot motion control. The generation process of the adaptive motion control command includes: when the robot is detected to be in a slope change section, adjusting the speed and torque output according to the slope; when the robot is detected to be in a curve, generating a smooth acceleration and deceleration command; when an obstacle is detected on the track ahead, generating a deceleration or stopping command; when a risk of slippage and derailment is detected, generating a braking command.

[0123] As described above, the computer-readable storage medium provided in this application embodiment, through a multi-sensor array mounted on the robot body, collects RFID absolute position, encoder relative position, IMU attitude data, and visual environment images in real time to form multi-source perception data. A data fusion algorithm is then used to perform spatiotemporal alignment and fusion processing on the multi-source information to generate the robot's current operating status information. Based on this operating status information, combined with task objectives and safety constraints, motion control commands are adaptively generated: dynamically adjusting speed and torque output to maintain uniform speed operation when the slope changes; automatically and smoothly decelerating when curves are encountered; performing deceleration, stopping, or obstacle avoidance operations when obstacles are detected; and performing braking operations when slipping. This improves the accuracy and efficiency of the track robot control.

[0124] Embodiments of this application also provide a computer program product capable of implementing all steps of the adaptive orbital robot motion control method with the execution subject being a server or client in the above embodiments. When executed by a processor, this computer program / instruction implements the steps of the adaptive orbital robot motion control method. For example, the computer program / instruction implements the following steps: Step S101: Deploy a multi-sensor array on the track robot body, collect multi-source perception data in real time during the operation of the track robot according to the multi-sensor array, perform spatiotemporal synchronization on the multi-source perception data, and determine the corresponding spatiotemporal aligned perception data. Step S102: Perform real-time quality assessment on the spatiotemporal alignment perception data, determine the confidence weight of each perception data, fuse the confidence weight with the spatiotemporal alignment perception data according to the preset extended Kalman filter, determine the corresponding robot fusion state vector, and determine the corresponding robot current running state according to the robot fusion state vector and the image data collected by the visual sensor, including the robot's current track slope state, curve state, obstacle state, and slippage / derailment state. Step S103: Based on the operating state, combined with the preset task objective and safety constraints, determine the corresponding adaptive motion control command, and send the motion control command to the robot execution layer to realize robot motion control. The generation process of the adaptive motion control command includes: when the robot is detected to be in a slope change section, adjusting the speed and torque output according to the slope; when the robot is detected to be in a curve, generating a smooth acceleration and deceleration command; when an obstacle is detected on the track ahead, generating a deceleration or stopping command; when a risk of slippage and derailment is detected, generating a braking command.

[0125] As described above, the computer program product provided in this application embodiment uses a multi-sensor array mounted on the robot body to collect RFID absolute position, encoder relative position, IMU attitude data, and visual environment images in real time, forming multi-source perception data. A data fusion algorithm is then used to perform spatiotemporal alignment and fusion processing on the multi-source information to generate the robot's current operating status information. Based on this operating status information, combined with task objectives and safety constraints, motion control commands are adaptively generated: dynamically adjusting speed and torque output to maintain uniform speed operation when the slope changes; automatically and smoothly decelerating when curves are encountered; performing deceleration, stopping, or obstacle avoidance operations when obstacles are detected; and performing braking operations when slipping. This improves the accuracy and efficiency of the track robot control.

[0126] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0127] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (devices), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations 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, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0128] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0129] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0130] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. An adaptive orbital robot motion control method, characterized in that, The method includes: A multi-sensor array is deployed on the body of the track robot. Multi-source perception data during the operation of the track robot is collected in real time based on the multi-sensor array. The multi-source perception data is spatiotemporally synchronized to determine the corresponding spatiotemporally aligned perception data. The spatiotemporal alignment perception data is evaluated in real time to determine the confidence weight of each perception data. The confidence weight is fused with the spatiotemporal alignment perception data according to a preset extended Kalman filter to determine the corresponding robot fusion state vector. Based on the robot fusion state vector and the image data collected by the vision sensor, the corresponding robot current running state is determined, including the robot's current track slope state, curve state, obstacle state, and slippage / derailment state. Based on the operating status, combined with the preset task objectives and safety constraints, corresponding adaptive motion control commands are determined and sent to the robot execution layer to achieve robot motion control. The generation process of the adaptive motion control commands includes: when the robot is detected to be in a slope change section, adjusting the speed and torque output according to the slope; when the robot is detected to be in a curve, generating smooth acceleration and deceleration commands; when an obstacle is detected on the track ahead, generating deceleration or stopping commands; and when a risk of slippage and derailment is detected, generating braking commands.

2. The adaptive orbital robot motion control method according to claim 1, characterized in that, The step of collecting multi-source perception data in real time during the operation of the orbital robot based on the multi-sensor array includes: The absolute position data of key nodes on the track is collected using RFID readers. The robot's relative position and speed data are collected based on the built-in photoelectric encoder in the motor; The robot's attitude data, including pitch and roll angles, are collected using a six-axis inertial measurement unit. Image data of the environment in front of the track is collected using a visual sensor.

3. The adaptive orbital robot motion control method according to claim 2, characterized in that, The real-time quality assessment includes: The confidence level of the data collected by the RFID reader is adjusted based on the track node tags, including assigning the highest confidence level to the absolute position data when it is determined that the RFID reader has read a tag. The confidence level of the data collected by the photoelectric encoder is adjusted based on the robot's slipping state, including assigning a first confidence level when there is no slipping and assigning a second confidence level when slipping is detected, wherein the second confidence level is lower than the first confidence level. The confidence level of the data collected by the six-axis inertial measurement unit is adjusted based on the robot's motion state, including assigning a third confidence level when the robot is stationary or moving at a constant speed, and assigning a fourth confidence level when the robot is vibrating violently. The fourth confidence level is lower than the third confidence level. The confidence level of the data collected by the visual sensor is adjusted based on the lighting conditions and image clarity. This includes assigning a fifth confidence level when the lighting is sufficient and the image is clear, and assigning a sixth confidence level when the lighting is insufficient or the image is blurry. The sixth confidence level is lower than the fifth confidence level.

4. The adaptive orbital robot motion control method according to claim 3, characterized in that, When the robot is detected to be in a slope change zone, the speed and torque output are adjusted according to the slope magnitude, including: When the robot enters the ramp and recognizes the ramp label, it determines the corresponding entry reference torque based on the preset ramp slope value, preset ramp speed curve and current load current value, so that the robot can smoothly enter the ramp at the preset target speed. When the robot enters the ramp and goes uphill, the pitch angle data of the robot body is obtained in real time from the six-axis inertial measurement unit as the current slope value. The current slope value, real-time speed value and load current value are input to the preset fuzzy PID controller. The PID parameters are self-tuned online through fuzzy rules to determine the corresponding target torque so that the robot can pass the ramp at a constant speed according to the target torque. When the robot enters the ramp and goes downhill, the corresponding braking force compensation value is determined based on the slope value, real-time speed value and load current value at the time of descent, and the braking force compensation value is added to the target torque to prevent overspeed.

5. The adaptive orbital robot motion control method according to claim 3, characterized in that, When the robot is detected to be on a curve, a smooth acceleration / deceleration command is generated, including: When the robot is detected to be at the entrance of a curve, the radius of curvature and length of the current curve are determined based on the robot's absolute position data and the preset track map information. The corresponding curve speed limit value is determined based on the radius of curvature. A smooth deceleration command is triggered before entering the curve based on the curve speed limit value, so that the robot enters the curve at a speed lower than that of the straight section. When the robot is detected to be on a curve, the roll angle data fed back by the six-axis inertial measurement unit is acquired in real time. When the roll angle data exceeds a preset threshold, a speed suppression command is generated to prevent overturning. When the robot is detected to be at the exit of a curve, a smooth acceleration command is generated to restore the cruising speed to that of the straight section.

6. The adaptive orbital robot motion control method according to claim 3, characterized in that, When an obstacle is detected on the track ahead, a deceleration or stopping command is generated, including: When an obstacle is detected on the track ahead, the image data collected by the vision sensor is used to infer the corresponding obstacle type, obstacle distance, and the area of ​​the track occupied by the obstacle through a locally deployed lightweight target detection model. Based on obstacle type and obstacle distance, generate tiered response commands, including: When the distance to the obstacle exceeds a first preset threshold, a deceleration command is generated; When the distance to the obstacle is less than or equal to the first preset threshold and greater than the second preset threshold, a parking instruction is generated and reported to the dispatch center. When the distance to the obstacle is less than or equal to the second preset threshold, an emergency braking command is generated, triggering a hard emergency stop. Based on the information about the area occupied by the obstacle on the track, determine whether it is possible to detour. If it is possible to detour, then determine the corresponding low-speed detour operation instruction according to the detour instruction issued by the dispatch center.

7. The adaptive orbital robot motion control method according to claim 3, characterized in that, When a risk of slippage and derailment is detected, a braking command is generated, including: The drive wheel speed fed back by the photoelectric encoder is compared with the robot body acceleration fed back by the six-axis inertial measurement unit. When the deviation between the rate of change of speed and the rate of change of acceleration exceeds a preset threshold, it is determined that there is a risk of slippage. In the state of slippage risk, a torque limiting command is generated to reduce the output torque of the drive wheel and an alarm message is generated and reported to the dispatch center. The roll angle fed back by the six-axis inertial measurement unit is compared with the preset track tilt angle model. When the roll angle deviates from the preset track attitude by more than a threshold, it is determined that there is a risk of derailment. In the case of derailment risk, an emergency stop command is immediately generated and the electromagnetic power failure brake is triggered to lock the drive wheels.

8. An adaptive orbital robot motion control device, characterized in that, The device includes: The multi-source data spatiotemporal alignment module is used to deploy a multi-sensor array on the body of the track robot, collect multi-source perception data in real time during the operation of the track robot according to the multi-sensor array, perform spatiotemporal synchronization on the multi-source perception data, and determine the corresponding spatiotemporal aligned perception data. The track robot state judgment module is used to perform real-time quality assessment on the spatiotemporal alignment perception data, determine the confidence weight of each perception data, fuse the confidence weight with the spatiotemporal alignment perception data according to a preset extended Kalman filter, determine the corresponding robot fusion state vector, and determine the corresponding robot current running state according to the robot fusion state vector and the image data collected by the visual sensor, including the current track slope state, curve state, obstacle state and slippage / derailment state of the robot. The adaptive motion control command generation module is used to determine the corresponding adaptive motion control command based on the operating state, combined with preset task objectives and safety constraints, and send the motion control command to the robot execution layer to realize robot motion control. The generation process of the adaptive motion control command includes: when the robot is detected to be in a slope change section, adjusting the speed and torque output according to the slope; when the robot is detected to be in a curve, generating a smooth acceleration or deceleration command; when an obstacle is detected on the track ahead, generating a deceleration or stopping command; and when a risk of slippage and derailment is detected, generating a braking command.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the adaptive orbital robot motion control method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the steps of the adaptive orbital robot motion control method according to any one of claims 1 to 7.

Images