Mobile device escape method and device, computer device and readable storage medium

Abstract

The invention discloses a mobile equipment escape method and device, computer equipment and a readable storage medium, relates to the technical field of smart homes, and aims to avoid falling into a dead zone due to global planning in a complex environment and improve the escape success rate while random blind collision is not needed, the number of attempts required for escape and collision loss are reduced and the equipment is prevented from being out of limit. And by directly determining and executing a clear escape action, the action torsional pendulum phenomenon caused by dynamically removing obstacles and repeatedly re-planning a path is avoided, so that the escape process is more coherent and stable, the secondary collision risk of the equipment in the escape process is reduced, the structure and performance of the equipment are protected, and the service life of the equipment is prolonged. And the service life and the working performance of equipment are prevented from being influenced. The method comprises the steps that when it is detected that the mobile device is in dilemma, a to-be-executed dilemma-escaping action is determined according to the current environment where the mobile device is located in a map; and controlling the mobile device to execute a dilemma-escaping action so as to enable the mobile device to be dilemma-escaping.

Description

Technical Field

[0001] This application relates to the field of smart home technology, and in particular to a method, apparatus, computer device, and readable storage medium for rescuing a mobile device from a difficult situation. Background Technology

[0002] In recent years, self-moving smart home devices have been moving towards full automation and intelligence, such as lawnmowers and robot vacuums. Taking a lawnmower as an example, to achieve full-coverage mowing, it typically first constructs a map of the work area, clearly defining boundaries, restricted areas, and obstacle locations, and then operates according to a planned zigzag path, ideally completing mowing efficiently. However, in actual zigzag mowing, due to limited positioning and tracking control accuracy, the lawnmower's actual position is prone to deviating from the map markings, causing it to collide with boundaries, restricted areas, or obstacles. After a collision, the lawnmower intersects with the obstacle, the original path planning fails, and the lawnmower gets stuck, requiring an extrication process to resume normal operation.

[0003] In related technologies, the lawnmower's collision sensors can be used to randomly collide with obstacles to escape, or the lawnmower can combine map and location information to find passable areas and replan its path to escape, or the lawnmower can clear obstacles from the map and replan its path to escape.

[0004] However, the applicant recognizes that the relevant technology has at least the following technical problems in its implementation: When using collision sensors to randomly collide and escape from trouble, multiple collisions may be required to successfully get out of trouble, which is inefficient, costly, and the lawnmower may also run out of the boundary. The method of combining map and location information to find passable areas and replan the path is computationally intensive and can easily cause the lawnmower to get into dead zones, resulting in a low success rate of getting out of trouble. The method of clearing obstacles from the map and replanning the path causes the lawnmower to wobble during the escape process, making it easy for the lawnmower to collide again, damaging the machine and affecting its service life and working performance. Summary of the Invention

[0005] In view of this, this application provides a method, apparatus, computer device and readable storage medium for rescuing a mobile device from a stuck situation, with the main purpose of improving the success rate of lawnmowers getting out of trouble.

[0006] According to a first aspect of this application, a method for escaping a trapped mobile device is provided, the method comprising: When a mobile device is detected to be in trouble, an escape action is determined based on the current environment of the mobile device on the map. The escape action includes at least one of a straight movement and a rotation movement. Control the mobile device to perform the escape action so that the mobile device can get out of trouble.

[0007] According to a second aspect of this application, a device for escaping a trapped mobile device is provided, the device comprising: The action determination module is used to determine the escape action to be performed based on the current environment of the mobile device in the map when the mobile device is detected to be in trouble. The escape action includes at least one of the following: a straight movement and a rotation movement. An action execution module is used to control the mobile device to perform the escape action so that the mobile device can get out of trouble.

[0008] According to a third aspect of this application, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described in any of the first aspects above.

[0009] According to a fourth aspect of this application, a readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in any one of the first aspects above.

[0010] By utilizing the above technical solutions, this application provides a method, apparatus, computer device, and readable storage medium for rescuing a mobile device from distress. When a mobile device is detected to be in distress, this application determines the rescuing action to be performed based on the current environment of the mobile device on the map, and controls the mobile device to execute the rescuing action to extricate the mobile device from the distress. By perceiving the current environment of the mobile device and making a purposeful rescuing decision, this application avoids random blind collisions, reduces the number of attempts and collision damage required for rescuing, prevents the device from going out of bounds, and avoids getting stuck in dead zones due to global planning in complex environments, thereby improving the success rate of rescuing. Furthermore, by directly determining and executing a specific rescuing action, this application avoids the swaying phenomenon caused by dynamically clearing obstacles and repeatedly replanning the path, making the rescuing process more coherent and stable, reducing the risk of secondary collisions during the rescuing process, thereby protecting the structure and performance of the device and avoiding affecting its service life and working performance.

[0011] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0012] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 This illustration shows a flowchart of a method for getting a mobile device out of trouble according to an embodiment of this application; Figure 2 A schematic diagram of a method for freeing a mobile device from a predicament, according to an embodiment of this application, is shown. Figure 3 A schematic diagram of a method for freeing a mobile device from a predicament, according to an embodiment of this application, is shown. Figure 4 A schematic diagram of a method for freeing a mobile device from a predicament, according to an embodiment of this application, is shown. Figure 5 A schematic diagram of a method for freeing a mobile device from a predicament, according to an embodiment of this application, is shown. Figure 6 A schematic diagram of a method for freeing a mobile device from a predicament, according to an embodiment of this application, is shown. Figure 7 This illustration shows a structural schematic diagram of a mobile device escape device provided in an embodiment of this application; Figure 8 A schematic diagram of the device structure of a computer device provided in an embodiment of this application is shown. Detailed Implementation

[0013] Exemplary embodiments of the present application will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this application will be thorough and complete, and will fully convey the scope of the present application to those skilled in the art.

[0014] This application provides a method for rescuing a mobile device from a difficult situation, such as... Figure 1 As shown, the method includes: S10: When a mobile device is detected to be in distress, determine the escape action to be performed based on the current environment of the mobile device on the map.

[0015] In this embodiment, the mobile device is a device capable of autonomous movement and completing designated tasks during movement. For example, in the smart home field, it could be a lawnmower, sweeper, or floor scrubber. For ease of explanation, the following description of the embodiments will use a lawnmower as an example. The lawnmower adopts a modular design. The power system is driven by a motor, which rotates the bottom blade disc via a gear transmission mechanism. The blade disc can use detachable blade sets to support replacement and height adjustment to adapt to different grass conditions. The positioning and navigation module integrates GPS, IMU inertial measurement unit, and UWB ultra-wideband positioning technology, combined with a top-mounted lidar to achieve 360° environmental perception, enabling real-time map building and planning of the optimal mowing path. Multiple infrared collision sensors and ultrasonic obstacle avoidance modules are distributed around the body, and a pressure sensor is equipped at the bottom to identify obstacles, boundaries, and terrain undulations. Combined with the AI ​​decision-making algorithm built into the central processor, it can dynamically adjust its operating posture or trigger preset escape actions, ensuring stable and efficient operation of the device in complex scenarios.

[0016] The escape maneuver includes at least one of two actions: a straight-line movement and a rotational movement. During the operation of a mobile device, various sensors, such as collision sensors, boundary sensors, and positioning sensors, monitor its operational status in real time. If the mobile device is detected to be in a difficult situation, such as encountering an obstacle and being unable to move forward, or getting stuck in a narrow area, the sensors immediately transmit signals to the device's central processing unit (CPU). The CPU then determines an appropriate escape maneuver, which includes at least one of two actions: a straight-line movement (moving forward or backward a certain distance) or a rotational movement (rotating clockwise or counterclockwise by a certain angle).

[0017] In this way, by perceiving and analyzing the current environment, purposeful escape decisions can be made, avoiding random and blind collisions, reducing the number of attempts required for escape, and minimizing losses caused by multiple collisions. It also effectively prevents the mobile device from going out of its working boundaries. Compared to traditional global planning escape methods, it avoids getting stuck in dead zones in complex environments due to improper global planning, greatly improving the success rate of escape. For example, suppose the mobile device is a lawnmower. The lawnmower is mowing in a garden and encounters a large rock. The lawnmower's collision sensor detects the collision and transmits the signal to the central processing unit (CPU). The CPU, combined with map information, determines that the space around the rock is relatively open and decides to execute an escape action of moving forward a certain distance, allowing the lawnmower to cross the rock and continue mowing.

[0018] S20: Control the mobile device to perform an escape action to free the mobile device from the predicament.

[0019] Once the escape maneuver is determined, corresponding control commands can be sent to the mobile device's drive system to help it escape the predicament. Furthermore, during the escape maneuver, the mobile device's sensors continuously monitor its operating status and changes in the surrounding environment. If any abnormalities are detected during the escape maneuver, such as encountering a new obstacle, the sensors will promptly provide feedback, allowing adjustments or replanning of the escape maneuver based on the feedback signal.

[0020] In this way, by directly determining and executing specific escape actions, the swaying phenomenon caused by dynamically clearing obstacles and repeatedly replanning the path is avoided. This reduces the risk of secondary collisions during the escape process, effectively protecting the structure and performance of the mobile equipment and preventing the impact on its lifespan and working performance caused by frequent collisions and unstable movements. For example, continuing with the example of the lawnmower encountering a rock, after determining to execute an escape action of moving forward a certain distance, a command is sent to the lawnmower's drive system. The drive system controls the motor to rotate, causing the wheels to turn, making the lawnmower move forward in a straight line and over the rock.

[0021] The method provided in this application, when a mobile device is detected to be in trouble, determines the escape action to be performed based on the current environment of the mobile device on the map, and controls the mobile device to perform the escape action to get the mobile device out of trouble. By perceiving the current environment of the mobile device, a purposeful escape decision is made. This avoids random blind collisions, reduces the number of attempts and collision damage required for escape, prevents the device from going out of bounds, and avoids getting stuck in dead zones due to global planning in complex environments, thus improving the success rate of escape. Furthermore, by directly determining and executing a clear escape action, the method avoids the swaying phenomenon caused by dynamically clearing obstacles and repeatedly replanning the path, making the escape process more coherent and stable, reducing the risk of secondary collisions during the escape process, protecting the structure and performance of the device, and avoiding affecting the service life and working performance of the device.

[0022] Optionally, in an embodiment of this application, before determining the escape action to be performed based on the current environment of the mobile device in the map when the mobile device is detected to be in trouble, the method further includes: when the mobile device is performing a work task, if the body of the mobile device is detected to be stuck in an obstacle in the map and / or the mobile device is physically stuck, the method determines that the mobile device is in trouble.

[0023] In this embodiment, to achieve more accurate and efficient extrication of the mobile device, multiple sensors built into the device monitor its operating status in real time and from all angles during the device's task execution. Among these, a positioning sensor continuously tracks the mobile device's location information within a pre-built map, while collision sensors, pressure sensors, and sensors for detecting wheel status work in concert. If the mobile device becomes stuck in an obstacle on the map, such as a crevice between rocks or bushes, the collision and pressure sensors immediately detect abnormal pressure changes and collision signals, and then relay these signals back.

[0024] Additionally, when a mobile device becomes physically stuck, it can also be considered to be in a predicament. Physical sticking includes at least one of the following: wheel slippage, front wheel jamming, or wheel entanglement. Wheel slippage refers to insufficient friction between the wheel and the ground, causing the wheel to spin freely. Sensors detecting wheel speed and rotation status will identify a discrepancy between the actual distance traveled and the expected distance, thus determining wheel slippage. Front wheel jamming refers to the mobile device's front wheel suddenly stopping due to being blocked by a large obstacle during movement; relevant sensors will promptly report this status. Wheel entanglement occurs when the mobile device encounters long grass, threads, etc., which become entangled on the wheel during operation, affecting normal wheel rotation; sensors detecting wheel rotation resistance can detect this anomaly. In this embodiment, the presence of any one or more of the above-mentioned situations—the device getting stuck in an obstacle or physically stuck—confirms that the mobile device is in a predicament.

[0025] In this way, through the collaborative work of multiple sensors, various situations in which mobile devices are stuck can be comprehensively and accurately identified, avoiding the misjudgments or omissions that may occur with a single sensor. This greatly improves the accuracy and reliability of trouble detection. Moreover, timely and accurate trouble detection is the foundation for effective subsequent extrication. Only by accurately determining that a mobile device is stuck can the appropriate extrication action be quickly determined based on its current environment on the map, reducing the time the mobile device spends in trouble, improving work efficiency, and reducing the risk of damage to the equipment caused by prolonged stuckness. For example, in a garden full of rocks and weeds, a lawnmower is performing a mowing task. When the lawnmower travels to a large rock, its front wheel gets stuck in the gap between the rock and the ground, causing the front wheel to stop. At this moment, the sensor detecting the rotation status of the front wheel immediately detects the abnormality and sends back an abnormality signal; at the same time, because the front wheel is stuck, the lawnmower body collides with the rock, and the collision sensor also detects the collision signal and sends back a feedback. Combining this information, it can be determined that the lawnmower is stuck.

[0026] In this embodiment of the application, optionally, the process of determining whether the mobile device is physically stuck is as follows: during the process of the mobile device performing a work task, the rotation speed of the wheels of the mobile device is acquired in real time, and the actual change value of a specified device parameter of the mobile device within a specified time threshold is recorded, wherein the specified device parameter is at least one of the device angle and the device movement distance; based on the wheel rotation speed, the theoretical change value of the specified device parameter within the specified time threshold is calculated; if the theoretical change value is greater than the actual change value and the difference between the two is greater than a preset difference threshold, it is determined that the mobile device is physically stuck, and the slip direction is determined based on the actual change value.

[0027] In this embodiment, the slippage causing physical jamming is actually divided into two types: rotational slippage and forward / backward slippage. To accurately identify these two types of slippage, multiple data acquisition tasks are performed simultaneously during the mobile device's task execution. Firstly, wheel speed sensors (ODOs) installed on each wheel acquire the wheel rotation speed in real time at a high frequency. The ODOs possess high sensitivity and high resolution, capable of capturing the wheel's rotational state at every moment, providing reliable basic data for subsequent analysis and judgment. Secondly, multiple built-in sensors record the actual changes in specified device parameters of the mobile device within a specified time threshold. These specified device parameters can be at least one of the following: device angle and device travel distance. The device angle can be measured using an inertial measurement unit (IMU) sensor, which can sense the mobile device's attitude changes in three-dimensional space in real time to determine the device angle. The device travel distance can be calculated using the mobile device's positioning system, combined with data from the wheel speed sensors and internal algorithms, ensuring the accuracy of the recorded actual changes.

[0028] In practical applications, when wheels slip during rotation, the rotational speed of one wheel will be significantly greater than that of the other due to differences in friction and other factors on both sides. For forward and backward slippage, ODO (Optical Direction Alignment) and Real-Time Dynamic Differential Positioning (RTK) technologies are primarily used. RTK technology provides high-precision positioning information, and combined with the wheel rotation measurements from ODO, a comprehensive assessment can be made to determine whether the mobile device is slipping in the forward or backward direction. Therefore, after obtaining the wheel rotation speed, a differential speed model can be used to calculate the theoretical changes in specified device parameters within a specified time threshold. For device angles, the differential speed model can calculate an approximate angle within the specified time threshold as the theoretical change value; for device travel distance, the differential speed model can calculate an approximate travel distance within the specified time threshold as the theoretical change value. Next, the calculated theoretical change value is compared with the actual recorded change value. When the theoretical change value is greater than the actual change value, and the difference between the two is greater than the preset difference threshold, it indicates that the theoretical change value is significantly greater than the actual change value, and it can be determined that the mobile device has been physically stuck. The preset difference threshold is a standard for determining whether the theoretical change value is significantly greater than the actual change value, derived from a large number of experiments and data analysis. It fully considers the normal error range of the theoretical change value under different working environments. Only when the difference exceeds this range does it indicate that the movement of the theoretical change value has been abnormally hindered, that is, physical stuck has occurred.

[0029] Furthermore, the direction of slippage is determined based on the actual change value, which is either left-handed or right-handed slippage. For example, if the angle change is clockwise and matches the characteristics of left-handed slippage, it can be determined as left-handed slippage. In addition, to improve the accuracy of the judgment, in actual application, the calculated wheel speed from the controller can be compared with the calculated value from the ODO measurement. This helps to confirm whether the judgment is accurate. The wheel speed calculation result from the controller is based on a more complex algorithm and global data, and it is cross-referenced with the calculated value from the ODO measurement, which can further eliminate misjudgments caused by individual sensor errors or local data anomalies.

[0030] In this way, by acquiring and comprehensively analyzing various key data in real time, physical jamming problems that occur during the operation of mobile devices can be detected in a timely manner, avoiding damage to the equipment caused by prolonged jamming and extending the service life of the equipment. Moreover, accurate jamming judgment and slippage direction determination provide a precise basis for the subsequent extrication strategy, enabling the most effective extrication actions to be taken according to the actual situation of the mobile device, improving extrication efficiency and success rate, reducing the need for manual intervention, and lowering usage costs.

[0031] In this embodiment of the application, optionally, when a mobile device is detected to be in a predicament, an escape action to be performed is determined based on the current environment of the mobile device in the map, including: when the mobile device is detected to be in a predicament, controlling the mobile device to perform an emergency stop; determining a target obstacle in the current environment of the mobile device in the map based on the predicament, wherein when the predicament indicates that the body of the mobile device is stuck in an obstacle in the map, the obstacle into which the body of the mobile device is stuck is taken as the target obstacle, and when the predicament indicates that the mobile device is physically stuck, adding an obstacle in the map based on the body of the mobile device, and taking the added obstacle as the target obstacle; generating an escape action to be performed based on the target obstacle.

[0032] In this embodiment, when a mobile device is detected to be stuck, it needs to be controlled to perform an emergency stop to immediately prevent it from continuing to move in a direction that could worsen the situation. Therefore, the obstacle causing the mobile device to get stuck could be a map boundary, a virtual area (a user-drawn no-go zone, prohibiting vehicles from entering), or a real obstacle. Real obstacles are more dangerous, potentially damaging the machine or colliding with pedestrians, so the mobile device needs to stop immediately. Specifically, a zero-speed signal can be sent to the underlying control motor of the mobile device to perform an emergency stop, preventing further damage to the device due to continuous movement and reducing the risk of further collisions or friction with the surrounding environment, creating a relatively stable initial state for subsequent extrication operations. For example, when a lawnmower encounters a tree stump hidden in the grass and gets stuck, the emergency stop can quickly stop the lawnmower from rotating, preventing the wheels from generating high temperatures due to continuous friction with the tree stump and damaging the surrounding vegetation.

[0033] Next, based on the specific predicament the mobile device is in and its current environment on the map, the target obstacle needs to be determined. As described above, there are two scenarios where the mobile device is in a predicament. The first scenario indicates that the lawnmower's body is stuck in an obstacle on the map, such as a tree stump. In this case, the obstacle into which the lawnmower's body is stuck will be directly used as the target obstacle. The second scenario indicates that the lawnmower is physically stuck, such as its wheels slipping and preventing it from moving forward or backward. In this case, a virtual obstacle needs to be added to the map, using the mobile device's body as a reference. This added obstacle will be used as the target obstacle to more accurately simulate the restricted environment the mobile device is currently in, allowing for the generation of more effective escape actions later.

[0034] Once the target obstacle is identified, a get-out-of-trouble maneuver can be generated based on it, aiming to extricate the mobile device from the predicament with minimal energy consumption and in the shortest possible time. For example, if the target obstacle is a small tree stump and the lawnmower is only partially stuck, a get-out-of-trouble maneuver can be generated that involves reversing, rotating, and then moving forward.

[0035] Thus, in this embodiment, by firstly using emergency braking and identifying the target obstacle, the predicament faced by the mobile device can be quickly and accurately identified, providing a clear direction and basis for subsequent escape operations, greatly improving the efficiency and success rate of escape; secondly, the escape actions generated based on the target obstacle fully consider the actual performance of the mobile device and the surrounding environmental factors, which can avoid secondary damage to the device due to blind operation, and also reduce the damage to the surrounding environment.

[0036] In this embodiment of the application, optionally, obstacles are added to the map based on the body of the mobile device, including: obtaining the slip direction when the mobile device is physically stuck; adding obstacles around the mobile device in the map with reference to the slip direction, wherein the added obstacles are cleared from the map when the mobile device completes its current task.

[0037] In this embodiment, the slip direction often indicates the presence of factors hindering the normal movement of the device in that direction. These could be uneven ground, hidden obstacles, etc. By adding obstacles around the device in the corresponding slip direction on the map, the restricted environment the device is currently in can be simulated, providing a more accurate reference for subsequent escape actions and path planning. Therefore, when adding virtual obstacles, it is necessary to obtain the slip direction when the mobile device physically gets stuck, and add obstacles around the mobile device on the map based on this slip direction. Specifically, in practical applications, if the device slips forward, a circular obstacle is added in front of the mobile device; if it slips backward, a circular obstacle is added behind the mobile device; if it slips to the left, a rectangular obstacle is added to the left side of the mobile device; and if it slips to the right, a rectangular obstacle is added to the right side of the mobile device. For example, see... Figure 2 , Figure 2 The green box represents the lawnmower. Currently, the lawnmower's front is facing downwards and to the right. Assuming slippage is detected in front of and to the left of the lawnmower, then... Figure 2 As shown, add a circular obstacle in front of the vehicle body within the green box, and add a rectangular obstacle to the left side of the vehicle body within the green box.

[0038] Furthermore, when a mobile device completes its current task, the added obstacles need to be cleared from the map. This is because, firstly, these obstacles are added to address temporary difficulties encountered during the current task and are generated based on specific work scenarios and environments. Therefore, once the task is completed, these temporarily added obstacles may no longer be applicable to subsequent tasks. For example, during a lawnmower task, the lawnmower slips on a wet, muddy area, and an obstacle is added to the map corresponding to that muddy location. However, by the next lawnmower task, this muddy area may have dried and hardened, no longer hindering the lawnmower's operation. Clearing the previously added obstacles allows the lawnmower to plan a more reasonable path in subsequent tasks, improving efficiency. Secondly, timely clearing of added obstacles avoids redundancy and confusion in map information. If these temporarily added obstacles remain on the map, the number of obstacles will increase as tasks progress, potentially leading to misjudgments during path planning by the mobile device and affecting its normal operation.

[0039] In this embodiment of the application, optionally, an escape action to be performed is generated based on the target obstacle, including: identifying the orientation information of the target obstacle relative to the mobile device in the map; generating a straight-line action based on the orientation information, and / or sampling safety points with the current position of the mobile device as the center to generate a rotation action; and combining the straight-line action and / or the rotation action to obtain the escape action to be performed.

[0040] In this embodiment, it is necessary to identify the location information of the target obstacle relative to the mobile device in the map. Specifically, this can be achieved by analyzing the surrounding environment information collected in real time by the sensors of the mobile device and combining it with map data to determine the location of the target obstacle relative to the lawnmower, such as whether it is located in front of, behind, to the left or to the right of the lawnmower.

[0041] After acquiring the orientation information, a straight-line motion is generated based on this information, causing the mobile device to move in a straight line in a relatively safe direction that is conducive to escaping the predicament. Simultaneously, a rotational motion can be generated by sampling a safety point centered on the mobile device's current location. This safety point sampling ensures that the mobile device will not collide with other obstacles during rotation. Based on the escape and parking points determined by the safety point sampling, a corresponding rotational motion is generated, causing the mobile device to rotate in a direction that allows it to successfully reach that parking point.

[0042] Finally, the straight-line and rotational actions are combined to obtain the escape action to be executed. It's important to note that generating the escape action is not a simple superposition of actions, but rather an optimization and sequence adjustment of the straight-line and rotational actions based on the actual predicament of the mobile device and the characteristics of the surrounding environment, ensuring the continuity and effectiveness of the escape action. For example, if the straight-line action is to move the lawnmower backward 30 centimeters, and the rotational action is to rotate the lawnmower 90 degrees clockwise, a complete escape action command can be generated based on the sequence and execution time of these two actions. This command would first move the lawnmower backward 30 centimeters, and then immediately rotate it 90 degrees clockwise, thus escaping the current predicament.

[0043] In this way, by accurately identifying the location information of the target obstacle, a clear direction for the mobile device to get out of trouble can be provided, avoiding blind operation that would worsen the predicament. Furthermore, by determining the straight-line direction and distance, as well as the stopping point for getting out of trouble and generating a rotational motion, the lawnmower can be ensured to move in the safest and most efficient way during the process of getting out of trouble, reducing the risk of collision with obstacles and protecting the equipment itself from damage. In addition, combining the straight-line motion and the rotational motion can generate a coherent and reasonable get-out-of-trouble action, improving the success rate and efficiency of getting out of trouble.

[0044] In this embodiment of the application, optionally, generating a straight-line action based on the orientation information includes: determining a straight-line direction based on the orientation information, wherein when the orientation information indicates that the target obstacle is in front of the mobile device, the direction in which the mobile device moves backward is taken as the straight-line direction, and when the orientation information indicates that the target obstacle is behind the mobile device, the direction in which the mobile device moves forward is taken as the straight-line direction; constructing a drivable area based on the current position of the mobile device on the map; querying a preset safe backward distance in the drivable area when the straight-line direction indicates backward movement, and taking the safe backward distance as the straight-line distance; and determining the first safe point in the drivable area in the direction in which the mobile device moves forward, and taking the distance between the current position of the mobile device and the first safe point as the straight-line distance when the straight-line direction indicates forward movement.

[0045] In this embodiment, the direction of travel needs to be determined first based on the acquired orientation information. The direction of travel includes forward and backward travel. Specifically, when the orientation information indicates that the target obstacle is in front of the mobile device, it means that the mobile device is directly facing the obstacle hindering its progress. Continuing to move forward may worsen the predicament or even damage the device. Therefore, the direction in which the mobile device moves backward is taken as the direction of travel. Conversely, when the orientation information indicates that the target obstacle is behind the mobile device, it means that the mobile device may be restricted from moving forward due to some reason. In this case, the direction in which the mobile device moves forward is taken as the direction of travel.

[0046] After determining the straight-ahead direction, the next step is to construct a drivable area based on the mobile device's current location on the map. This ensures the mobile device can only perform extrication operations within this drivable area, preventing it from crossing boundaries. Next, the straight-ahead distance is determined. Specifically, when the straight-ahead direction indicates reversing, a preset safe reversing distance needs to be queried within the drivable area and used as the straight-ahead distance. This preset safe reversing distance is a safety-compliant value derived from extensive experiments and data analysis, comprehensively considering the mobile device's own performance (such as braking distance and size), spatial limitations of the surrounding environment, and potential risks. In practical applications, for lawnmowers, this preset safe reversing distance can be set to 10 centimeters, and the lawnmower should be instructed to reverse according to this standard. For example, as... Figure 3 As shown in the figure, the blue box represents the lawnmower, and the green circular area represents the obstacle. Assuming that it is safe to move 0.2m behind the lawnmower, the movement mode of the lawnmower is set to backward, that is, backward, and the straight-line distance is set to the safe backward distance of 10 centimeters, that is, backward_dist = 0.1m.

[0047] When the direction of travel is indicated as straight ahead, the first safe point needs to be determined within the drivable area in the direction the mobile device is traveling. The distance between the current position of the mobile device and the first safe point is taken as the straight-line distance. Determining the first safe point requires real-time scanning and analysis of the environment in the direction the mobile device is traveling. This can be achieved by using sensors on the mobile device itself (such as lidar or ultrasonic sensors) to detect the area ahead, collecting information such as the distance and shape of surrounding objects. Based on this information, combined with preset safety standards (such as the minimum safe distance from obstacles), a point is found in the direction of travel that ensures the mobile device's safe passage without creating new obstacles—this is the first safe point. For example, as shown... Figure 4 As shown in the figure, the blue box represents the lawnmower, and the green quadrilateral area represents the obstacle. When it is safe in front of the lawnmower, the lawnmower's movement mode is set to forward, that is, to move forward to disengage and move forward a distance equal to the distance between the current vehicle direction and the first safe point in this direction, forward_dist.

[0048] In this way, by clearly defining the straight-line direction corresponding to different directional information, mobile devices can quickly and correctly react when faced with various difficulties, avoiding the worsening of the predicament due to blind actions. Furthermore, by querying the preset safe retreat distance and determining the first safe point to calculate the straight-line distance, the safety of mobile devices during the escape process can be ensured, the risk of collision with obstacles can be reduced, and the device itself can be protected from damage, which helps to improve the efficiency and success rate of escape.

[0049] In this embodiment of the application, optionally, a drivable area is constructed based on the current location of the mobile device on the map, including: obtaining the original map constructed by the mobile device for the area where it is currently performing a task, and obtaining the boundary information associated with the area where the mobile device is currently performing a task; performing semantic recognition on the boundary labels corresponding to the boundary boundaries of each area included in the boundary information to determine the traversable area among all the area boundaries included in the boundary information; and integrating the area indicated by the original map and the traversable area to obtain the drivable area.

[0050] In this embodiment, the initial map constructed by the mobile device for the area where it is currently performing the lawnmowing task is first acquired. This initial map is the result of the mobile device's preliminary mapping and modeling of the work area based on its sensor data, such as GPS positioning and LiDAR scanning, reflecting the mobile device's basic understanding of the work area. Next, to more accurately define the mobile device's traversable range, boundary information associated with the current area needs to be acquired. This boundary information can come from various sources, including but not limited to user-manually set electronic fences, geographic boundary data provided by a Geographic Information System (GIS), or boundary features extracted from satellite images or images taken by drones using image recognition technology. The boundary information includes a detailed description of each area's boundary and a corresponding boundary label for each boundary. These labels identify the nature of the boundary, such as whether it is permissible to cross or whether it is a restricted area.

[0051] After acquiring boundary information, semantic recognition technology can be used to deeply analyze these boundary labels. Through natural language processing or machine learning algorithms, the textual information in the boundary labels can be transformed into machine-understandable semantic information, thereby accurately identifying traversable areas among all regional boundaries. For example, for lawnmowers, some boundaries may be marked as "touchable boundaries" or "vehicles are allowed to cross the boundary," meaning that lawnmowers can safely cross these boundaries while performing their tasks without triggering safety alarms or causing damage. In practical applications, semantic recognition can employ techniques such as OST (Object-Semantic Transformation) to rotate or adjust the representation of boundaries, more accurately matching the driving logic and safety requirements of mobile devices.

[0052] Finally, the area indicated by the original map is integrated with the boundary-crossing area determined through semantic recognition to generate a comprehensive map of the drivable area. This map not only includes the work area that the mobile device needs to cover, but also clearly marks which boundaries can be safely crossed and which need to be strictly observed. This improves the autonomous navigation capability and work efficiency of the mobile device, because it can flexibly plan the driving path of the mobile device without violating safety rules, avoid obstacles, and ensure the comprehensive completion of the work task.

[0053] Optionally, taking the current location of the mobile device as the center, safety point sampling is performed to generate a rotation action, including: taking the current location of the mobile device as the center, emitting a ray as a sampling path at specified step angles within a preset radius until the maximum angle indicated by the preset angle range is reached, thus obtaining multiple sampling paths; performing safety point sampling based on the multiple sampling paths to obtain multiple candidate safety points, and determining the safety point information corresponding to each candidate safety point, wherein the safety point information includes the rotation angle, movement distance, and safety length of the corresponding candidate safety point; evaluating the cost value of each candidate safety point based on the safety point information corresponding to each candidate safety point to obtain the safety point cost value corresponding to each candidate safety point; selecting the candidate safety point with the lowest safety point cost value among the multiple candidate safety points as the escape parking point, and generating a rotation action using the rotation angle included in the safety point information corresponding to the escape parking point.

[0054] In this embodiment, firstly, safety point sampling is performed within a preset radius using the current location of the mobile device as the center. The preset radius can be determined comprehensively based on factors such as the size of the mobile device, its maximum rotation radius, and the distribution of common obstacles. For example, for a lawnmower with a diameter of 80 centimeters, considering the space required for rotation, the preset radius might be set to 1.2 meters to ensure that a sufficient number of potential safe areas are covered during sampling. After determining the radius, the mobile device emits a ray as the sampling path at specified step angles. In this embodiment, the specified step angle can be... This means that the angle can be set to 30 degrees. The mobile device will start from 0 degrees and emit a ray every 30 degrees until it reaches the maximum angle indicated by the preset angle range. Assuming the preset angle range is from 0 degrees to 360 degrees, this will result in 12 sampling paths.

[0055] Next, based on these multiple sampling paths, the mobile device begins sampling for safe points, obtaining multiple candidate safe points. During the sampling process, the mobile device can utilize its onboard sensors, such as LiDAR and ultrasonic sensors, to perform real-time environmental detection along each sampling path. When a sensor detects that there are no obstacles around a certain location or that the distance to an obstacle meets a preset safe distance requirement, that location is marked as a candidate safe point. Simultaneously, the mobile device also determines the safety point information corresponding to each candidate safe point, including the rotation angle, movement distance, and safe length. The rotation angle refers to the angle that the mobile device needs to rotate from its current position to a direction that allows it to face the candidate safe point. The rotation angle of each candidate safe point can be used It indicates that the moving distance refers to the distance that a movable device moves in a straight line from its current location to the candidate safe point. The movement distance of each candidate safe point can be used The safe length refers to the length of space reserved in front of the candidate safe point if the mobile device reaches it, sufficient to ensure the safe stopping and subsequent operation of the mobile device. The safety length of a candidate safe point can be used as follows: This is represented as follows. For example, if a lawnmower detects a candidate safety point on a sampling path, and calculations show that rotating 45 degrees from the current position towards that point and moving 60 centimeters in a straight line to that point, with no obstacles 20 centimeters in front of the point, then the corresponding rotation angle for this candidate safety point is 45 degrees, the movement distance is 60 centimeters, and the safety length is 20 centimeters. It should be noted that in practical applications, candidate safety points can be numbered according to the order in which they are collected. For example, multiple collected candidate safety points can be represented as follows: , to And generate a candidate point set that includes multiple candidate safe points. So that in the subsequent candidate point set Select the optimal candidate safe point for escape. See details in [link to relevant documentation]. Figure 5 , Figure 5 The blue box represents a mobile device. Using the current position of the blue box as the center, a ray is emitted every 30 degrees as the sampling path, thus obtaining... Figure 5 The green lines indicate 12 sampling paths; then, safe points are traversed along each sampling path to obtain candidate safe points. to .

[0056] Then, based on the safety point information corresponding to each candidate safety point, the cost value of each candidate safety point is evaluated to obtain the safety point cost value corresponding to each candidate safety point. Cost value evaluation is a complex process that comprehensively considers multiple factors; it aims to assess the merits of each candidate safety point as an escape and parking point. In this embodiment, the cost value considers the magnitude of the rotation angle, the length of the movement distance, and the adequacy of the safety length. A smaller rotation angle indicates a smaller range of rotation required by the mobile device, making operation simpler and resulting in a relatively lower cost value. A shorter movement distance means less energy consumed by the mobile device, higher escape efficiency, and a correspondingly lower cost value. A more ample safety length ensures higher safety for the mobile device after parking, enabling subsequent operations (such as adjusting direction or continuing to drive), and a lower cost value. Therefore, a cost value formula can be used to comprehensively evaluate the magnitude of the rotation angle, the length of the movement distance, and the adequacy of the safety length of each candidate safety point to obtain the safety point cost value corresponding to each candidate safety point.

[0057] Finally, among multiple candidate safe points, the one with the lowest safety point cost is selected as the escape and parking point. This ensures that the mobile device can extricate itself from the predicament in the optimal way, considering both ease of operation and energy consumption and safety. Furthermore, after determining the escape and parking point, the rotation angle included in the corresponding safe point information needs to be used to generate a rotational motion, preparing for subsequent movement and escape.

[0058] In this way, by comprehensively detecting multiple sampling paths, it is possible to cover the possible safe areas in all directions around the mobile device, increasing the probability of finding a suitable parking spot. Moreover, by comprehensively considering factors such as rotation angle, movement distance and safe length for cost evaluation, the optimal parking spot can be selected, making the lawnmower easier to operate, consuming less energy and being safer during the process of getting out of trouble.

[0059] It should be noted that in practical applications, safety point sampling can begin from the current direction of travel of the mobile device, so that the mobile device can move forward to complete the escape; or, if an obstacle is detected in front of the mobile device and the mobile device needs to move backward to complete the escape, safety point sampling can also begin from the current backward direction of the mobile device, so that the mobile device can move backward to complete the escape. This application does not specifically limit the direction on which safety point sampling is based.

[0060] In this embodiment of the application, optionally, safety point sampling is performed based on multiple sampling paths to obtain multiple candidate safety points, including: sampling safety points on each sampling path to obtain multiple initial safety points; performing passability detection on each initial safety point; and determining multiple candidate safety points among the multiple initial safety points based on the results of the passability detection, wherein passability detection includes at least one of reachability detection and detection of the presence of other obstacles during the driving process.

[0061] In this embodiment, safety points are first sampled along each sampling path according to preset sampling intervals or conditions. Each sampling point represents a possible safe stopping or turning point for the mobile device at that location, thus obtaining multiple initial safety points. Considering that not all initial safety points truly meet the safe driving requirements of the mobile device, a passability test is then performed on each initial safety point. The test process includes, but is not limited to, reachability detection, i.e., verifying whether the mobile device can reach the initial safety point from its current location or a known safety point via an unobstructed path; and the detection of other obstacles during the journey, i.e., simulating whether the mobile device will collide with any known or unknown obstacles during its journey from its current location to the initial safety point. It should be noted that different passability tests can be performed individually or in combination to ensure the comprehensiveness and accuracy of the test results.

[0062] Finally, based on the results of the passability detection, multiple candidate safe points that meet the conditions for safe driving are selected from multiple initial safe points. These candidate safe points not only provide mobile devices with more reliable options for stopping and turning, but also greatly enhance the mobile devices' ability to cope with and flexibility in complex environments.

[0063] In this embodiment of the application, optionally, determining the safety point information corresponding to each candidate safety point includes: for each candidate safety point, determining the target sampling path where the candidate safety point is located among multiple sampling paths; calculating the rotation angle of the candidate safety point based on the number of specified step angles traversed when emitting the ray of the target sampling path; counting the distance between the current location of the mobile device and the candidate safety point as the moving distance, and identifying the safety distance reserved in front of the candidate safety point in the map as the safety length; and combining the rotation angle, moving distance, and safety length to obtain the safety point information corresponding to the candidate safety point.

[0064] In this embodiment, for each candidate safety point, the target sampling path where the candidate safety point is located is first determined from multiple sampling paths. The rotation angle of the candidate safety point is then calculated based on the number of specified step angles traversed when the ray of the target sampling path is emitted. For example, if the specified step angle is set to 30 degrees, and the lawnmower starts emitting a ray from 0 degrees, but the ray corresponding to the target sampling path is emitted after traversing 3 step angles (i.e., rotating 90 degrees (3 × 30 degrees) from 0 degrees), then the rotation angle of this candidate safety point is 90 degrees. Furthermore, in practical applications, if the lawnmower's current orientation deviates from the initially set 0-degree direction, the angle needs to be corrected according to the actual situation to ensure that the calculated rotation angle allows the lawnmower to accurately face the candidate safety point, avoiding inaccurate rotation angles that could prevent successful arrival at the candidate safety point or collisions with other obstacles.

[0065] Next, the distance between the mobile device's current location and the candidate safe point is calculated as the movement distance. Specifically, this can be achieved using the mobile device's own sensors, such as a laser sensor. By measuring the time difference between laser emission and reception, and combining this with the speed of light, the straight-line distance between the lawnmower and the candidate safe point can be accurately calculated. Simultaneously, a safe distance must be identified in the map in front of the candidate safe point as the safe length. This can be determined by identifying the range within which there are no obstacles in front of the candidate safe point, and then determining the safe length based on this range. For example, if there are no obstacles within 20 centimeters in front of the candidate safe point, then the safe length is 20 centimeters.

[0066] Finally, the calculated rotation angle, movement distance, and safety length are combined to obtain the safety point information corresponding to the candidate safety point, so as to comprehensively and accurately describe the characteristics of each candidate safety point and provide a reliable basis for subsequent escape decisions.

[0067] Optionally, in this embodiment, the cost value of each candidate safety point is evaluated based on the safety point information corresponding to each candidate safety point, and the safety point cost value corresponding to each candidate safety point is obtained. This includes: querying the weights corresponding to the rotation angle, movement distance, and safety length respectively; using the queried weights to perform weight calculation on the safety point information corresponding to each candidate safety point, and using the calculated result as the safety point cost value corresponding to the corresponding candidate safety point.

[0068] In this embodiment, it is necessary to first query the weights corresponding to the rotation angle, travel distance, and safety length. When the movable device is a lawnmower, the weight of the rotation angle can be set considering the ease of operation and energy consumption. The smaller the rotation angle, the shorter the time required for the lawnmower to rotate, the lower the energy consumption, and the simpler the operation, resulting in less wear on the mechanical structure. Therefore, the weight of the rotation angle can be set relatively high to encourage the lawnmower to prioritize candidate safety points with smaller rotation angles. The weight of the travel distance is related to the lawnmower's endurance and escape efficiency. The shorter the travel distance, the less energy the lawnmower consumes, allowing it to complete more mowing tasks with limited battery power and reach the escape point faster, improving escape efficiency. Therefore, the weight of the travel distance can also be reasonably set to ensure that the lawnmower does not get stuck in a new predicament due to excessive travel distance when trying to escape a difficult situation. The weighting of the safety length needs to focus on the safety of the lawnmower after it has come out of trouble and stopped. The more safety length there is, the less likely the lawnmower will collide with surrounding obstacles when it performs subsequent operations (such as adjusting direction or continuing to drive) after stopping. This provides a safer operating space for the lawnmower. Therefore, the weighting of the safety length can be set reasonably to ensure the safe operation of the lawnmower after it has come out of trouble.

[0069] After retrieving the weights corresponding to the rotation angle, movement distance, and safety length, these weights can be used to calculate the safety point value for each candidate safety point. By appropriately setting the weights for the rotation angle, movement distance, and safety length, and using these weights for weighted calculations, multiple factors affecting the escape efficiency can be comprehensively considered. This allows for a comprehensive and accurate evaluation of the merits of each candidate safety point as an escape parking point, selecting the optimal escape parking point for the mobile device, improving escape efficiency and safety, and ensuring stable operation of the mobile device even in complex environments.

[0070] In this embodiment of the application, optionally, controlling the mobile device to perform an escape action to get the mobile device out of trouble includes: sending a preset straight-line speed to the mobile device to control the mobile device to move in the straight-line direction indicated by the straight-line action, and controlling the mobile device to stop moving forward when the distance moved by the mobile device reaches the straight-line distance indicated by the straight-line action; and / or sending a preset rotation speed to the mobile device to control the mobile device to rotate, and controlling the mobile device to stop rotating when the angle of rotation of the mobile device reaches the rotation angle indicated by the rotation action.

[0071] In this embodiment, during the straight-line movement in the escape maneuver, a preset straight-line speed can be sent to the mobile device to control it to move in the straight-line direction indicated by the straight-line movement. During the movement, the sensors on the mobile device monitor its distance in real time. When the distance traveled by the mobile device reaches the straight-line distance indicated by the straight-line movement, the mobile device needs to be stopped. In practical applications, the preset straight-line speed can be 0.2 m / s. When controlling the mobile device to move forward in a straight line, a speed of 0.2 m / s can be sent to the controller of the mobile device, and when the positioning data calculates that the mobile device has moved forward_dist, a speed of 0 is sent to make the mobile device stop abruptly. Conversely, when controlling the mobile device to move backward in a straight line, a speed of -0.2 m / s can be sent to the controller of the mobile device, and when the positioning data calculates that the mobile device has moved backward_dist, a speed of 0 is sent to make the mobile device stop abruptly.

[0072] In addition to straight-line movement, the escape mechanism also includes rotation. During rotation, a preset rotation speed can be sent to the mobile device to control its rotation. During rotation, the mobile device's sensors monitor the rotation angle in real time. When the detected rotation angle reaches the specified rotation angle, the device stops rotating. In practical applications, the preset rotation speed can be 0.4 rad / s. When the rotation angle is greater than 0, a speed of 0.4 rad / s can be sent to the mobile device, and when the rotation angle reaches the specified angle, a speed of 0 is sent to stop the device. Conversely, when the rotation angle is less than 0, a speed of -0.4 rad / s can be sent, and when the rotation angle reaches the specified angle, a speed of 0 is sent to stop the device.

[0073] In practical applications, escape maneuvers can include one or both of the following: linear movement and rotational movement. These two movements can be used individually or in combination, depending on the specific situation. For example, see... Figure 6 The green box represents a lawnmower. The lawnmower is stuck between the obstacles represented by the three yellow circles. In this situation, you can first control the lawnmower to rotate, adjust its direction, and then move it straight to get out of trouble. This allows you to provide precise escape instructions for mobile devices based on different situations, improving the success rate and efficiency of getting out of trouble.

[0074] In this embodiment of the application, the method may optionally further include: continuously detecting whether the mobile device is physically stuck during the process of controlling the mobile device to perform the escape action; when the mobile device is detected to be physically stuck, stopping the escape action, re-determining a new escape action to be performed based on the current environment of the mobile device in the map, and controlling the mobile device to perform the new escape action so that the mobile device can get out of trouble.

[0075] In this embodiment, while controlling the mobile device to perform the escape action, it is necessary to continuously and in real time monitor the motion status, position changes and surrounding environment information of the mobile device through various sensors on it, such as wheel speed sensors, accelerometers, tilt sensors and vision or lidar, in order to determine whether the mobile device is still in a physically stuck state, that is, whether it has failed to escape the predicament as expected.

[0076] Once a physical jam is detected in a mobile device, the current escape attempt must be stopped to avoid wasting energy and causing further damage to the device. Next, the mobile device's current environment on the map is re-analyzed, and new escape actions are determined. The mobile device is then controlled to execute these new actions. Through more precise and flexible operation, the mobile device can more effectively overcome its current predicament and successfully escape the jammed state.

[0077] Optionally, in this embodiment of the application, the method further includes: detecting the current operating status of the cutterhead of the mobile device when the escaping action is detected to be completed; controlling the mobile device to continue performing the current work task when the cutterhead of the mobile device is detected to be currently operating normally; and determining a new escaping action to be performed based on the current environment of the mobile device in the map, and controlling the mobile device to perform the new escaping action so that the mobile device can escape the predicament.

[0078] In this embodiment, after the mobile device completes a series of escape maneuvers, it is not immediately instructed to continue its original task. Instead, the system first initiates a status check on the critical components of the mobile device. In the case of a lawnmower, the critical component is the blade, which, as the core working part, directly affects the efficiency and quality of the mowing task. Therefore, a comprehensive monitoring of the blade's current operating status can be achieved through a built-in sensor network, such as motor speed sensors and vibration sensors. These sensors can capture key parameters in real time, including blade rotation speed, vibration frequency, and any abnormal noises, providing accurate information for determining whether the blade is operating normally.

[0079] Accordingly, if the test results show that the cutter head is currently operating normally—that is, the rotation speed is stable, the vibration is within a reasonable range, and there are no abnormal noises—the mobile equipment can be controlled to continue performing its current work task, such as continuing to mow along the preset path. This ensures that the mobile equipment can quickly resume work after being freed from its predicament, improving work efficiency. However, if the cutter head is detected to be malfunctioning—such as abnormal rotation speed, excessive vibration, or obvious malfunction noises—it may mean that although the mobile equipment has been freed from its predicament, the distance it has moved is insufficient for the cutter head to operate normally. Therefore, it is necessary to determine a new set of extrication actions based on the current environment of the lawnmower on the map, including obstacle distribution and terrain features, so that the mobile equipment can move to a position where the cutter head can operate normally, allowing it to continue performing its current work task. This ensures the continuity and efficiency of the mobile equipment's operation and provides strong support for its stable operation.

[0080] Furthermore, as Figure 1 To specifically implement the method, this application provides a mobile device escape device, such as... Figure 7 As shown, the device includes: an action determination module 701 and an action execution module 702.

[0081] The action determination module 701 is used to determine the escape action to be performed based on the current environment of the mobile device in the map when the mobile device is detected to be in trouble. The escape action includes at least one of a straight movement and a rotation movement. The action execution module 702 is used to control the mobile device to perform the escape action so that the mobile device can get out of trouble.

[0082] In specific application scenarios, the device further includes: The distress detection module is used to determine that the mobile device is in a distress when the body of the mobile device is stuck in an obstacle in the map and / or the mobile device is physically stuck during the performance of the work task. The physical distress includes at least one of wheel slippage, front wheel jamming, and wheel entanglement.

[0083] In specific application scenarios, the trouble detection module is also used to acquire the wheel rotation speed of the mobile device in real time during the mobile device's work task, and record the actual change value of a specified device parameter of the mobile device within a specified time threshold, wherein the specified device parameter is at least one of device angle and device movement distance; calculate the theoretical change value of the specified device parameter within the specified time threshold based on the wheel rotation speed; and determine that the mobile device has physically gotten stuck when the theoretical change value is greater than the actual change value and the difference between the two is greater than a preset difference threshold, and determine the slip direction based on the actual change value.

[0084] In a specific application scenario, the action determination module 701 is used to control the mobile device to perform an emergency stop when it is detected that the mobile device is in a predicament; based on the predicament of the mobile device, it determines a target obstacle in the environment where the mobile device is currently located on the map, wherein when the predicament indicates that the body of the mobile device is stuck in an obstacle in the map, the obstacle into which the body of the mobile device is stuck is taken as the target obstacle; and when the predicament indicates that the mobile device is physically stuck, an obstacle is added to the map based on the body of the mobile device, and the added obstacle is taken as the target obstacle; based on the target obstacle, the action to be executed is generated.

[0085] In a specific application scenario, the action determination module 701 is used to obtain the slip direction when the mobile device is physically stuck; in the map, referring to the slip direction, the obstacle is added around the mobile device; and when the mobile device completes its current task, the added obstacle is cleared from the map.

[0086] In a specific application scenario, the action determination module 701 is used to identify the orientation information of the target obstacle relative to the mobile device in the map; generate the straight-line action based on the orientation information, and / or perform safety point sampling with the current position of the mobile device as the center to generate the rotation action; and combine the straight-line action and / or the rotation action to obtain the escape action to be executed.

[0087] In a specific application scenario, the action determination module 701 is used to determine the straight-line direction based on the orientation information. Specifically, when the orientation information indicates that the target obstacle is in front of the mobile device, the direction in which the mobile device moves backward is taken as the straight-line direction; and when the orientation information indicates that the target obstacle is behind the mobile device, the direction in which the mobile device moves forward is taken as the straight-line direction. A drivable area is constructed based on the current location of the mobile device on the map. When the straight-line direction indicates reversal, a preset safe reversal distance is queried within the drivable area, and this safe reversal distance is taken as the straight-line distance. When the straight-line direction indicates forward movement, the first safe point is determined within the drivable area in the direction the mobile device is moving forward, and the distance between the current location of the mobile device and the first safe point is taken as the straight-line distance.

[0088] In a specific application scenario, the action determination module 701 is used to obtain the original map constructed by the mobile device for the area where it is currently performing a work task, and to obtain the boundary information associated with the area where the mobile device is currently performing a work task; to perform semantic recognition on the boundary labels corresponding to each area boundary included in the boundary information, so as to determine the traversable area among all area boundaries included in the boundary information; and to integrate the area indicated by the original map and the traversable area to obtain the drivable area.

[0089] In a specific application scenario, the action determination module 701 is used to take the current position of the mobile device as the center, and within a preset radius, emit a ray at specified step angles as sampling paths until the maximum angle indicated by the preset angle range is reached, thus obtaining multiple sampling paths; based on the multiple sampling paths, safety point sampling is performed to obtain multiple candidate safety points, and safety point information corresponding to each candidate safety point is determined, wherein the safety point information includes the rotation angle, movement distance, and safety length of the corresponding candidate safety point; according to the safety point information corresponding to each candidate safety point, the cost value of each candidate safety point is evaluated to obtain the safety point cost value corresponding to each candidate safety point; among the multiple candidate safety points, the candidate safety point with the lowest safety point cost value is selected as the escape parking point, and the rotation action is generated using the rotation angle included in the safety point information corresponding to the escape parking point.

[0090] In a specific application scenario, the action determination module 701 is used to sample safety points on each of the sampling paths to obtain multiple initial safety points; to perform passability detection on each initial safety point; and to determine multiple candidate safety points among the multiple initial safety points based on the results of the passability detection, wherein the passability detection includes at least one of reachability detection and detection of the presence of other obstacles during the driving process.

[0091] In a specific application scenario, the action determination module 701 is used to determine the target sampling path where the candidate safety point is located among the multiple sampling paths for each candidate safety point; calculate the rotation angle of the candidate safety point based on the number of specified step angles traversed when the ray of the target sampling path is emitted; count the distance between the current position of the mobile device and the candidate safety point as the moving distance, and identify the safe distance reserved in front of the candidate safety point in the map as the safe length; combine the rotation angle, the moving distance, and the safe length to obtain the safety point information corresponding to the candidate safety point.

[0092] In a specific application scenario, the action determination module 701 is used to query the weights corresponding to the rotation angle, the movement distance, and the safety length respectively; use the queried weights to perform weight calculation on the safety point information corresponding to each candidate safety point, and use the calculated result as the safety point value corresponding to the corresponding candidate safety point.

[0093] In specific application scenarios, the action execution module 702 is used to send a preset straight-line speed to the mobile device to control the mobile device to move in the straight-line direction indicated by the straight-line action, and to control the mobile device to stop moving forward when it is detected that the distance moved by the mobile device reaches the straight-line distance indicated by the straight-line action; and / or, to send a preset rotation speed to the mobile device to control the mobile device to rotate, and to control the mobile device to stop rotating when it is detected that the rotation angle of the mobile device reaches the rotation angle indicated by the rotation action.

[0094] In specific application scenarios, the action determination module 701 is also used to continuously detect whether the mobile device is physically stuck during the process of controlling the mobile device to perform the escape action; when the mobile device is detected to be physically stuck, the escape action is stopped, a new escape action is determined based on the current environment of the mobile device in the map, and the mobile device is controlled to perform the new escape action so that the mobile device can get out of trouble.

[0095] In specific application scenarios, the action determination module 701 is further configured to detect the current operating status of the cutterhead of the mobile device when the escaping action is detected to have been completed; when the cutterhead of the mobile device is detected to be operating normally, control the mobile device to continue performing the current work task; when the cutterhead of the mobile device is detected to be unable to operate normally, re-determine a new escaping action to be performed based on the current environment of the mobile device on the map, and control the mobile device to perform the new escaping action so that the mobile device can escape the predicament.

[0096] The device provided in this application, when detecting that a mobile device is in trouble, determines the escape action to be performed based on the current environment of the mobile device in the map, and controls the mobile device to perform the escape action so that the mobile device can get out of trouble. By perceiving the current environment of the mobile device, it makes a purposeful escape decision, which avoids random blind collisions, reduces the number of attempts and collision damage required for escape, prevents the device from going out of bounds, and avoids getting stuck in dead zones due to global planning in complex environments, thereby improving the success rate of escape. Moreover, by directly determining and executing a clear escape action, it avoids the swaying phenomenon caused by dynamically clearing obstacles and repeatedly replanning the path, making the escape process more coherent and stable, reducing the risk of secondary collisions during the escape process, protecting the structure and performance of the device, and avoiding affecting the service life and working performance of the device.

[0097] It should be noted that other corresponding descriptions of the functional units involved in the mobile device extrication device provided in this application embodiment can be found by referring to... Figures 1 to 6 The corresponding descriptions in [the document] will not be repeated here.

[0098] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0099] The above embodiments and the technical features in the embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0100] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

[0101] In an exemplary embodiment, see Figure 8 The invention also provides a computer device including a bus, a processor, a memory, and a communication interface. It may also include an input / output interface and a display device, wherein the various functional units can communicate with each other via the bus. The memory stores a computer program, and the processor executes the program stored in the memory to perform the mobile device extrication method described in the above embodiments.

[0102] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the mobile device extrication method.

[0103] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented in hardware or by using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0104] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this application.

[0105] Those skilled in the art will understand that the modules in the apparatus of the implementation scenario can be distributed within the apparatus of the implementation scenario as described, or they can be located in one or more apparatuses different from this implementation scenario, with corresponding changes. The modules of the above-described implementation scenario can be combined into one module, or they can be further divided into multiple sub-modules.

[0106] The serial numbers in this application are for descriptive purposes only and do not represent the superiority or inferiority of the implementation scenario.

[0107] The above disclosures are only a few specific implementation scenarios of this application. However, this application is not limited to these. Any variations that can be conceived by those skilled in the art should fall within the protection scope of this application.

Claims

1. A method for freeing a mobile device from a difficult situation, characterized in that, include: When a mobile device is detected to be in trouble, an escape action is determined based on the current environment of the mobile device on the map. The escape action includes at least one of a straight movement and a rotation movement. Control the mobile device to perform the escape action so that the mobile device can get out of trouble.

2. The method according to claim 1, characterized in that, Before determining the escape action to be performed based on the current environment of the mobile device on the map when it is detected that the mobile device is in distress, the method further includes: During the operation of the mobile device, when the mobile device is detected to be stuck in an obstacle on the map and / or the mobile device is physically stuck, it is determined that the mobile device is in trouble. The physical stuckness includes at least one of wheel slippage, front wheel jamming, and wheel entanglement.

3. The method according to claim 2, characterized in that, The method further includes: During the execution of the work task by the mobile device, the rotation speed of the wheels of the mobile device is acquired in real time, and the actual change value of the specified device parameter of the mobile device within a specified time threshold is recorded, wherein the specified device parameter is at least one of the device angle and the device moving distance. Based on the wheel rotation speed, calculate the theoretical change value of the specified equipment parameter within the specified time threshold; If the theoretical change value is greater than the actual change value and the difference between the two is greater than a preset difference threshold, it is determined that the mobile device is physically stuck, and the slip direction is determined based on the actual change value.

4. The method according to claim 1, characterized in that, When a mobile device is detected to be in distress, the step of determining the escape action to be performed based on the current environment of the mobile device on the map includes: When a mobile device is detected to be in distress, the mobile device is controlled to perform an emergency stop braking. Based on the predicament the mobile device is in, a target obstacle is determined in the environment where the mobile device is currently located on the map. Specifically, when the predicament indicates that the mobile device is stuck in an obstacle on the map, the obstacle into which the mobile device is stuck is taken as the target obstacle. When the predicament indicates that the mobile device is physically stuck, an obstacle is added to the map based on the mobile device's body, and the added obstacle is taken as the target obstacle. Based on the target obstacle, the escape action to be performed is generated.

5. The method according to claim 4, characterized in that, The step of adding obstacles to the map based on the body of the mobile device includes: Obtain the slip direction when the mobile device becomes physically stuck; In the map, obstacles are added around the mobile device according to the slip direction. When the mobile device completes its current task, the added obstacles are cleared from the map.

6. The method according to claim 4, characterized in that, The step of generating the escape action to be performed based on the target obstacle includes: Identify the orientation information of the target obstacle relative to the mobile device in the map; Based on the orientation information, the linear motion is generated, and / or the rotational motion is generated by sampling safety points with the current location of the mobile device as the center. The straight-line action and / or the rotational action are combined to obtain the escape action to be performed.

7. The method according to claim 6, characterized in that, The step of generating the linear motion based on the orientation information includes: Based on the orientation information, the straight-ahead direction is determined, wherein when the orientation information indicates that the target obstacle is in front of the mobile device, the direction in which the mobile device moves backward is taken as the straight-ahead direction, and when the orientation information indicates that the target obstacle is behind the mobile device, the direction in which the mobile device moves forward is taken as the straight-ahead direction. Based on the current location of the mobile device on the map, a drivable area is constructed; When the straight-ahead direction indicates reversing, a preset safe reversing distance is queried in the drivable area, and the safe reversing distance is used as the straight-ahead distance; When the direction of travel is indicated, a first safe point is determined in the drivable area in the direction of travel of the mobile device, and the distance between the current position of the mobile device and the first safe point is taken as the straight-line distance.

8. The method according to claim 7, characterized in that, The step of constructing a drivable area based on the current location of the mobile device on the map includes: Obtain the original map constructed by the mobile device for the area where the current task is being performed, and obtain the boundary information associated with the area where the mobile device is currently performing the task; Semantic recognition is performed on the boundary labels corresponding to each region boundary included in the boundary information to determine the crossable regions among all region boundaries included in the boundary information; The drivable area is obtained by integrating the area indicated by the original map and the traversable area.

9. The method according to claim 6, characterized in that, The step of sampling a safe point using the current position of the mobile device as the center to generate the rotation action includes: Using the current location of the mobile device as the center, within a preset radius, a ray is emitted at specified step angles as a sampling path until the maximum angle indicated by the preset angle range is reached, thus obtaining multiple sampling paths; Based on the multiple sampling paths, safety points are sampled to obtain multiple candidate safety points, and safety point information corresponding to each candidate safety point is determined. The safety point information includes the rotation angle, movement distance, and safety length of the corresponding candidate safety point. Based on the security point information corresponding to each candidate security point, the cost value of each candidate security point is evaluated to obtain the security point cost value corresponding to each candidate security point. Among the multiple candidate safety points, the candidate safety point with the lowest safety point value is selected as the escape and parking point. The rotation action is generated using the rotation angle included in the safety point information corresponding to the escape and parking point.

10. The method according to claim 9, characterized in that, The process of sampling security points based on the multiple sampling paths yields multiple candidate security points, including: Safety points are sampled on each of the sampling paths to obtain multiple initial safety points; For each of the initial safe points, a passability detection is performed, and based on the results of the passability detection, a plurality of candidate safe points are determined from the plurality of initial safe points, wherein the passability detection includes at least one of reachability detection and detection of the presence of other obstacles during the journey.

11. The method according to claim 9, characterized in that, The step of determining the security point information corresponding to each candidate security point includes: For each candidate safe point, determine the target sampling path where the candidate safe point is located from the plurality of sampling paths; The rotation angle of the candidate safety point is calculated based on the number of specified step angles traversed when the ray of the target sampling path is emitted. The distance between the current location of the mobile device and the candidate safe point is calculated as the moving distance, and the safe distance reserved in front of the candidate safe point is identified in the map as the safe length; The rotation angle, the movement distance, and the safety length are combined to obtain the safety point information corresponding to the candidate safety point.

12. The method according to claim 9, characterized in that, The step of evaluating the cost value of each candidate security point based on the security point information corresponding to each candidate security point, and obtaining the security point cost value corresponding to each candidate security point, includes: Query the weights corresponding to the rotation angle, the movement distance, and the safety length respectively; The weights obtained from the query are used to calculate the weights of the security point information corresponding to each candidate security point, and the calculated results are used as the security point value corresponding to the candidate security point.

13. The method according to claim 1, characterized in that, The control of the mobile device to perform the escape action to free the mobile device from the predicament includes: A preset straight-line speed is sent to the mobile device to control it to move in the straight-line direction indicated by the straight-line action, and when the distance traveled by the mobile device reaches the straight-line distance indicated by the straight-line action, the mobile device is controlled to stop moving forward; and / or, A preset rotation speed is sent to the mobile device to control it to rotate. When the angle of rotation of the mobile device reaches the rotation angle indicated by the rotation action, the mobile device is controlled to stop rotating.

14. The method according to claim 1, characterized in that, The method further includes: During the process of controlling the mobile device to perform the escape action, the system continuously monitors whether the mobile device is physically stuck. When the mobile device is detected to be physically stuck, the escape action is stopped, and a new escape action is determined based on the current environment of the mobile device on the map. The mobile device is then controlled to execute the new escape action so that it can get out of trouble.

15. The method according to claim 1, characterized in that, The method further includes: Upon detecting that the escape action has been completed, the current operating status of the cutterhead of the mobile device is detected; When it is detected that the cutter head of the mobile device is currently operating normally, the mobile device is controlled to continue performing the current work task; When it is detected that the cutterhead of the mobile device is not functioning properly, a new escape action is determined based on the current environment of the mobile device on the map, and the mobile device is controlled to execute the new escape action so that the mobile device can get out of trouble.

16. A device for freeing a mobile device from a difficult situation, characterized in that, include: The action determination module is used to determine the escape action to be performed based on the current environment of the mobile device in the map when the mobile device is detected to be in trouble. The escape action includes at least one of the following: a straight movement and a rotation movement. An action execution module is used to control the mobile device to perform the escape action so that the mobile device can get out of trouble.

17. The apparatus according to claim 16, characterized in that, The device further includes: The distress detection module is used to determine that the mobile device is in a distress when the body of the mobile device is stuck in an obstacle in the map and / or the mobile device is physically stuck during the performance of the work task. The physical distress includes at least one of wheel slippage, front wheel jamming, and wheel entanglement.

18. The apparatus according to claim 17, characterized in that, The trouble detection module is further configured to acquire the wheel rotation speed of the mobile device in real time during the mobile device's work task, and record the actual change value of a specified device parameter of the mobile device within a specified time threshold, wherein the specified device parameter is at least one of device angle and device movement distance; calculate the theoretical change value of the specified device parameter within the specified time threshold based on the wheel rotation speed; and determine that the mobile device has physically gotten stuck when the theoretical change value is greater than the actual change value and the difference between the two is greater than a preset difference threshold, and determine the slip direction based on the actual change value.

19. The apparatus according to claim 16, characterized in that, The action determination module is used to control the mobile device to perform an emergency stop when it detects that the mobile device is in a predicament; based on the predicament, it determines a target obstacle in the current environment of the mobile device on the map, wherein when the predicament indicates that the mobile device is stuck in an obstacle on the map, the obstacle into which the mobile device is stuck is taken as the target obstacle; and when the predicament indicates that the mobile device is physically stuck, an obstacle is added to the map based on the mobile device, and the added obstacle is taken as the target obstacle; based on the target obstacle, the action to be executed is generated.

20. The apparatus according to claim 19, characterized in that, The action determination module is used to obtain the slip direction when the mobile device is physically stuck; in the map, referring to the slip direction, the obstacle is added around the mobile device, wherein the added obstacle is cleared in the map when the mobile device completes its current task.

21. The apparatus according to claim 19, characterized in that, The action determination module is used to identify the orientation information of the target obstacle relative to the mobile device in the map; generate the straight-line action based on the orientation information, and / or perform safety point sampling with the current position of the mobile device as the center to generate the rotation action; and combine the straight-line action and / or the rotation action to obtain the escape action to be executed.

22. The apparatus according to claim 21, characterized in that, The action determination module is used to determine the straight-line direction based on the orientation information, wherein when the orientation information indicates that the target obstacle is in front of the mobile device, the direction in which the mobile device moves backward is taken as the straight-line direction; and when the orientation information indicates that the target obstacle is behind the mobile device, the direction in which the mobile device moves forward is taken as the straight-line direction. A drivable area is constructed based on the current position of the mobile device on the map. When the straight-line direction indicates reversal, a preset safe reversal distance is queried within the drivable area, and this safe reversal distance is taken as the straight-line distance. When the straight-line direction indicates forward movement, the first safe point is determined within the drivable area in the direction the mobile device is moving forward, and the distance between the current position of the mobile device and the first safe point is taken as the straight-line distance.

23. The apparatus according to claim 22, characterized in that, The action determination module is used to obtain the original map constructed by the mobile device for the area where it is currently performing a task, and to obtain the boundary information associated with the area where the mobile device is currently performing a task; to perform semantic recognition on the boundary labels corresponding to each area boundary included in the boundary information, so as to determine the traversable area among all area boundaries included in the boundary information; and to integrate the area indicated by the original map and the traversable area to obtain the drivable area.

24. The apparatus according to claim 21, characterized in that, The action determination module is used to emit a ray as a sampling path within a preset radius, with the current position of the mobile device as the center, at specified step angle intervals, until the maximum angle value indicated by the preset angle range is reached, thus obtaining multiple sampling paths; based on the multiple sampling paths, safety point sampling is performed to obtain multiple candidate safety points, and safety point information corresponding to each candidate safety point is determined, wherein the safety point information includes the rotation angle, movement distance, and safety length of the corresponding candidate safety point; based on the safety point information corresponding to each candidate safety point, the cost value of each candidate safety point is evaluated to obtain the safety point cost value corresponding to each candidate safety point; among the multiple candidate safety points, the candidate safety point with the lowest safety point cost value is selected as the escape parking point, and the rotation action is generated using the rotation angle included in the safety point information corresponding to the escape parking point.

25. The apparatus according to claim 24, characterized in that, The action determination module is used to sample safety points on each of the sampling paths to obtain multiple initial safety points; For each of the initial safe points, a passability detection is performed, and based on the results of the passability detection, a plurality of candidate safe points are determined from the plurality of initial safe points, wherein the passability detection includes at least one of reachability detection and detection of the presence of other obstacles during the journey.

26. The apparatus according to claim 24, characterized in that, The action determination module is used to determine the target sampling path where the candidate safety point is located among the multiple sampling paths for each candidate safety point; and to calculate the rotation angle of the candidate safety point based on the number of specified step angles traversed when the ray of the target sampling path is emitted. The distance between the current location of the mobile device and the candidate safe point is calculated as the moving distance, and the safe distance reserved in front of the candidate safe point is identified in the map as the safe length; The rotation angle, the movement distance, and the safety length are combined to obtain the safety point information corresponding to the candidate safety point.

27. The apparatus according to claim 24, characterized in that, The action determination module is used to query the weights corresponding to the rotation angle, the movement distance, and the safety length, respectively. The weights obtained from the query are used to calculate the weights of the security point information corresponding to each candidate security point, and the calculated results are used as the security point value corresponding to the candidate security point.

28. The apparatus according to claim 16, characterized in that, The action execution module is configured to send a preset straight-line speed to the mobile device to control the mobile device to move in the straight-line direction indicated by the straight-line action, and to control the mobile device to stop moving forward when the distance moved by the mobile device reaches the straight-line distance indicated by the straight-line action; and / or, send a preset rotation speed to the mobile device to control the mobile device to rotate, and to control the mobile device to stop rotating when the angle of rotation of the mobile device reaches the rotation angle indicated by the rotation action.

29. The apparatus according to claim 16, characterized in that, The action determination module is further configured to continuously detect whether the mobile device is physically stuck during the process of controlling the mobile device to perform the escape action; when the mobile device is detected to be physically stuck, the escape action is stopped, a new escape action is determined based on the current environment of the mobile device in the map, and the mobile device is controlled to perform the new escape action so that the mobile device can get out of trouble.

30. The apparatus according to claim 16, characterized in that, The action determination module is further configured to detect the current operating status of the cutterhead of the mobile device when the escaping action is detected to have been completed; when the cutterhead of the mobile device is detected to be operating normally, control the mobile device to continue performing the current work task; when the cutterhead of the mobile device is detected to be unable to operate normally, re-determine a new escaping action to be performed based on the current environment of the mobile device on the map, and control the mobile device to perform the new escaping action so that the mobile device can escape the predicament.

31. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 15.

32. A readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 15.

Images