A method and system for cloud-based robot control

Abstract

This invention provides a method and system for cloud-controlled robots, relating to intelligent control technology. The robot terminal acquires environmental data and object state information based on a sensor module, and uploads this data to a cloud server via a communication module. The cloud server determines the object's specifications based on the object state information, calculates the gripping points based on these specifications, determines constraint segments based on the environmental data, calculates trajectory offset values ​​based on the object specifications and segment parameters, and generates control commands based on the gripping points and trajectory offset values, sending these commands to the robot terminal. The cloud server acquires real-time status data of the robot terminal while executing the control commands, and generates action adjustment information based on this real-time status data to update the robot terminal's commands. This enables more efficient and flexible object handling control of the robot.

Description

Technical Field

[0001] This invention relates to intelligent control technology, and more particularly to a method and system for controlling a robot via the cloud. Background Technology

[0002] In today's diverse and complex work scenarios, robots are widely used for material handling in various fields such as industrial production and logistics warehousing, and these scenarios have extremely high requirements for the efficiency and flexibility of robot handling.

[0003] Currently, in existing technologies, controlling robots to transport objects mainly relies on the robot terminal executing preset programs. In this mode, technicians typically pre-program detailed transport procedures, such as precise transport paths, accurate gripping positions, and precise placement positions, into the robot terminal's local control system. The robot terminal then performs the corresponding object transport operations according to the program written in the local control system. However, this traditional preset program control method has many undeniable drawbacks. The biggest problem is the severe lack of flexibility. Once the position or layout of objects in the working environment changes, or if objects of different sizes need to be transported, the program needs to be rewritten and debugged. This process is not only time-consuming but also requires a large amount of manpower for repeated debugging, resulting in low robot transport efficiency.

[0004] Therefore, how to achieve more efficient and flexible object handling control of robots has become an urgent problem to be solved. Summary of the Invention

[0005] This invention provides a method and system for cloud-based robot control, which enables more efficient and flexible object handling control of the robot.

[0006] A first aspect of the present invention provides a method for cloud-controlled robot, characterized in that it includes:

[0007] The robot terminal acquires environmental data and object status information based on the sensor module, and uploads the environmental data and object status information to the cloud server based on the communication module;

[0008] The cloud server determines the object's specifications based on its state information and calculates the grab points based on those specifications.

[0009] The cloud server determines the constrained road segment based on environmental data, calculates the trajectory offset value based on the object specifications and road segment parameters, and generates control commands based on the gripping point and trajectory offset value, which are then sent to the robot terminal.

[0010] The system acquires real-time status data of the robot terminal when executing control commands, and the cloud server generates action adjustment information based on the real-time status data to update the robot terminal's commands.

[0011] Optionally, in one possible implementation of the first aspect, the cloud server determines the object's dimensions based on the object's state information and calculates the grab points based on the object's dimensions, including:

[0012] The cloud server determines the object's dimensions based on image data from various perspectives, and the object's status information includes image data from each perspective.

[0013] Obtain the maximum reach of the robot terminal, determine the half-circumference based on the object's specifications, and when the maximum reach is greater than the half-circumference, determine the center point of the opposite side of the object as the gripping point.

[0014] When the maximum reach length is less than half the circumference, multi-robot collaboration is triggered to determine the transport surface corresponding to each of the multiple robot terminals. The point located at the center point gripping offset distance in the transport surface is determined as the gripping point corresponding to each robot terminal.

[0015] Optionally, in one possible implementation of the first aspect, multi-machine collaboration is triggered when the maximum arm extension length is less than half the circumference, determining the respective transport surfaces of multiple robot terminals, and determining the points on the transport surfaces located at the center point gripping offset distance as the gripping points of each robot terminal, including:

[0016] Determine the half-arm length of the robot terminal and obtain the grasping range corresponding to the half-arm length on the side of each object;

[0017] When the grasping range is smaller than the range threshold of the transport surface, the number of terminals for multi-machine collaboration is determined to be the preset number at the four corners, and the sides of adjacent objects are sequentially determined as the transport surfaces corresponding to the respective robot terminals.

[0018] When there is no grasping range smaller than the transport surface range threshold, the number of terminals for multi-machine collaboration is determined to be a pre-set number diagonally, and the diagonally adjacent object sides are determined to be the transport surface corresponding to the corresponding robot terminal.

[0019] The point located at the center point of the handling surface at the gripping offset distance is determined as the gripping point corresponding to each robot terminal.

[0020] Optionally, in one possible implementation of the first aspect, the grab offset distance is determined by the following steps:

[0021] The volume of an object is calculated based on its specifications, and an offset coefficient is obtained based on the ratio of the reference volume to the object's volume.

[0022] The capture offset distance is obtained by multiplying the offset coefficient and the reference offset distance and then performing a weighted summation.

[0023] Optionally, in one possible implementation of the first aspect, the cloud server identifies passable and impassable areas based on pixel classification results in environmental data, wherein the environmental data includes image data corresponding to the frontal view.

[0024] Obtain the centerline of the passable area. When the angle between the centerline of the path and the current travel direction of the robot terminal is greater than a threshold, retrieve the area map.

[0025] Determine the location of the robot terminal in the area map, and identify the road segments where the location point is located and which the robot terminal has not yet traversed as constraint road segments;

[0026] Obtain the turning angle of the constrained road segment, and calculate the turning speed of the robot terminal from the current driving direction to the direction corresponding to the turning angle based on the specifications of the item. The road segment parameters include the turning angle, and the trajectory offset value includes the turning speed.

[0027] Optionally, in one possible implementation of the first aspect, obtaining the turning angle of the constrained road segment includes:

[0028] Obtain the movement trajectory of the robot terminal in the area map within a preset time period, and determine the tangent direction of the end point of the movement trajectory as the driving direction of the robot terminal;

[0029] Determine the centerline of the constrained road segment. When the tangent directions of all path points on the centerline are consistent, determine the angle between the centerline and the driving direction as the turning angle, and determine the direction from the driving direction to the centerline as the turning direction.

[0030] When the tangent directions of the path points on the center line are inconsistent, the path point on the center line that is closest to the positioning point of the robot terminal is determined as the starting point.

[0031] Starting from the starting point, obtain path segments composed of path points with the same tangent direction as the starting point. Determine the angle between the path segment and the driving direction as the turning angle, and determine the direction from the driving direction to the path segment as the turning direction.

[0032] Optionally, in one possible implementation of the first aspect, calculating the turning speed of the robot terminal from the current driving direction to the direction corresponding to the turning angle based on the item specifications includes:

[0033] The volume of an object is calculated based on its specifications, and the first adjustment coefficient is obtained by weighting the reciprocal of the object's volume.

[0034] Obtain the item's attributes. When the item's attribute is fragile, obtain the preset coefficient corresponding to the risk level of the target item as the second adjustment coefficient.

[0035] The total adjustment coefficient is obtained by combining the first and second adjustment coefficients. The turning speed is obtained by multiplying the total adjustment coefficient and the reference speed.

[0036] Optionally, in one possible implementation of the first aspect, after the cloud server determines the constrained road segment based on environmental data, it further includes:

[0037] Obtain the width of the target item in the direction perpendicular to the current driving direction, and obtain the width of the constrained road segment;

[0038] When the width of an item is greater than or equal to the width of a road segment, a reference model corresponding to the item's specifications is generated. The reference model is then adjusted according to a preset rotation angle, and the reference width corresponding to the reference model is obtained.

[0039] The target angle is determined when the reference width is smaller than the road segment width. The target angle is then sent to the robot terminal to adjust the angle of the item.

[0040] Optionally, in one possible implementation of the first aspect, real-time status data of the robot terminal when executing control commands is obtained, and the cloud server generates motion adjustment information based on the real-time status data to update the robot terminal's commands, including:

[0041] Vibration values ​​are acquired by the sensor module of the robot terminal, and the real-time status data includes vibration values.

[0042] When the vibration value is greater than the vibration threshold, the vibration difference between the vibration value and the reference vibration value is obtained, and the speed adjustment value is obtained by weighted calculation of the ratio of the vibration difference to the reference difference.

[0043] The motion adjustment information is obtained by subtracting the speed adjustment value from the current speed of the robot terminal, and the speed of the robot terminal is updated based on the motion adjustment information.

[0044] A second aspect of the present invention provides a cloud-controlled robot system, comprising:

[0045] The upload module is used by the robot terminal to acquire environmental data and object status information based on the sensor module, and to upload the environmental data and object status information to the cloud server based on the communication module.

[0046] The determination module is used by the cloud server to determine the object's specifications based on the object's state information and to calculate the grab points based on the object's specifications.

[0047] The calculation module is used by the cloud server to determine the constraint road segment based on environmental data, calculate the trajectory offset value based on the object size and road segment parameters, and generate control commands based on the gripping point and trajectory offset value and send them to the robot terminal.

[0048] The generation module is used to acquire real-time status data of the robot terminal when executing control commands. The cloud server generates action adjustment information based on the real-time status data to update the robot terminal's commands.

[0049] The beneficial effects of this invention are as follows:

[0050] 1. This invention can centrally process and analyze environmental data and object status information uploaded by robot terminals through a cloud server, and issue precise control commands based on the analysis results, thereby achieving flexible and efficient handling control of the robot. Through the efficient collaboration between the cloud server and the robot terminal, the robot terminal can achieve more precise and flexible object handling operations.

[0051] 2. This invention can transmit environmental data and object status information to the cloud server through the communication module, ensuring that environmental data and object status information can be uploaded from the robot terminal to the cloud server in real time and reliably. This allows the cloud server to obtain the latest status of the robot's surrounding environment and objects in a timely manner. Furthermore, the cloud server can determine the object's specifications through object status information containing multi-view image data and calculate the optimal gripping point by combining it with the maximum arm span of the robot terminal. This ensures the stability and balance of the robot terminal when gripping objects, preventing objects from tilting or falling during transportation.

[0052] 3. In this invention, the cloud server can perform a comprehensive and detailed analysis of the robot terminal's surrounding environment based on real-time environmental data uploaded by the robot terminal, accurately determining constraint sections. By accurately identifying constraint sections, the robot terminal can prepare in advance, avoiding collisions caused by failure to detect constraint sections in time, thus improving the robot terminal's safety during object handling. After determining the constraint sections, the cloud server can calculate the trajectory offset value by comprehensively considering the object's specifications and the section parameters, thereby effectively reducing the risk of objects falling or being damaged during handling, and improving the robot terminal's efficiency in passing through constraint sections.

[0053] 4. In this invention, during the process of the robot terminal executing control commands to transport objects, its sensor module continuously collects real-time status data during the transport process and can continuously send its own real-time status data to the cloud server. The cloud server can analyze the received real-time status data, especially the vibration data. When abnormal vibration data is detected, the cloud server can generate corresponding action adjustment information. Based on the action adjustment information, the speed of the robot terminal can be updated accordingly, thereby improving the stability of the robot terminal when transporting objects. Attached Figure Description

[0054] Figure 1 This is a flowchart illustrating a cloud-based robot control method provided in an embodiment of the present invention;

[0055] Figure 2 This is a schematic diagram of determining the transport surface corresponding to a robot terminal according to an embodiment of the present invention;

[0056] Figure 3 This is a schematic diagram illustrating how to determine the turning direction according to an embodiment of the present invention;

[0057] Figure 4 This is a schematic diagram of the structure of a cloud-controlled robot system provided in an embodiment of the present invention. Detailed Implementation

[0058] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0059] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0060] See Figure 1 This is a schematic diagram of a cloud-based robot control method provided in an embodiment of the present invention. Figure 1 The execution entity of the method shown can be a software and / or hardware device. The execution entity of this application can include, but is not limited to, at least one of the following: user equipment, network equipment, etc. User equipment can include, but is not limited to, computers, smartphones, personal digital assistants (PDAs), and the aforementioned electronic devices. Network equipment can include, but is not limited to, a single network server, a server group consisting of multiple network servers, or a cloud based on cloud computing consisting of a large number of computers or network servers. Cloud computing is a type of distributed computing, consisting of a super virtual computer composed of a group of loosely coupled computers. This embodiment does not limit this. Steps S1 to S4 are detailed as follows:

[0061] S1, the robot terminal acquires environmental data and object status information based on the sensor module, and uploads the environmental data and object status information to the cloud server based on the communication module.

[0062] The robot terminal refers to the robotic device that performs specific tasks. It integrates sensor modules and communication modules, enabling it to perform object handling operations autonomously or remotely under control. The sensor module refers to the components on the robot terminal used to perceive the external environment, including but not limited to vision sensors. Environmental data refers to the spatial information of the robot's surrounding environment, and object status information refers to the all-round image information of the object to be handled, obtained by the vision sensors in the sensor module from multiple different perspectives. The communication module is the module that connects the robot terminal to the cloud server. Through this module, the robot terminal can reliably upload the environmental data and object status information collected by the sensor module to the cloud server in real time, and at the same time receive control commands issued by the cloud server. The cloud server is a central platform that can centrally process the data uploaded by the robot terminal. It has powerful computing and storage capabilities and can use advanced technologies such as big data analysis and machine learning to deeply analyze the received data, generate precise control commands, and send them to the robot terminal through the communication network.

[0063] In practical applications, robots have been deeply integrated into many key areas such as industrial production, logistics and distribution, medical services, and home-based elderly care. For example, in large logistics warehouses, massive amounts of goods flow continuously, requiring a large number of robots to work collaboratively. They not only need to accurately and efficiently complete core tasks such as handling and sorting goods in a complex environment where the stacking layout of goods is constantly changing, but also need to cleverly avoid interference from other operating equipment such as forklifts to ensure that the entire logistics process is smooth and unobstructed. Traditional robot control methods mainly rely on preset program control. In relatively fixed scenarios with little environmental change, such as the handling process on a production line for specific parts, robots can follow the preset program to systematically pick up objects from designated locations, move them along predetermined paths, and place them at the target location. However, once the position and layout of objects in the working environment change, or when it is necessary to handle objects of different sizes and shapes, the program must be rewritten and debugged. This process is time-consuming and labor-intensive. For example, in logistics warehouses, there are many types of goods and their stacking positions are constantly changing, making it difficult for robots relying on preset programs to adapt. This solution can centrally process and analyze the data uploaded by the robot terminal through a cloud server, and issue precise control commands based on the analysis results, thereby achieving flexible and efficient handling control of the robot. Through this efficient collaboration between the cloud server and the robot terminal, the robot can achieve more precise and flexible object handling operations, while also having stronger environmental adaptability and task execution efficiency.

[0064] Specifically, when a robot performs object handling tasks, its onboard sensor modules, such as vision sensors, can collect real-time environmental data. This environmental data refers to the spatial information of the robot's surroundings, including the location, shape, and distance of obstacles, as well as the distribution of passable areas. For example, during the handling process, the robot can scan its surroundings using vision sensors to obtain information such as whether there are obstacles ahead, whether the passage is clear, and the location of turning points. This environmental data is crucial for the robot to plan its handling path, avoid obstacles, and dynamically adjust its movement trajectory. By collecting environmental data in real time, it can accurately determine whether the robot needs to turn or adjust its speed, thereby improving the handling efficiency. The task's safety and efficiency are ensured, and the sensor module can capture images of the object to be transported from multiple different perspectives. By capturing images of the object from various angles, comprehensive image information of the object to be transported can be obtained, i.e., object status information. This image information covers all sides, top, and bottom of the object. After acquiring environmental data and object status information, the robot terminal can upload this data to the cloud server through its communication module. The communication module typically supports high-speed, low-latency data transmission protocols to ensure that environmental data and object status information can be transmitted to the cloud in real time and reliably. Environmental data and object status information can provide the cloud server with sufficient information for subsequent data analysis.

[0065] The above implementation method ensures that environmental data and object status information can be transmitted to the cloud server in real time and reliably.

[0066] S2: The cloud server determines the object's specifications based on the object's status information and calculates the grab points based on the object's specifications.

[0067] After receiving environmental data and object status information uploaded by the robot terminal, the cloud server can analyze the received data. Specifically, by analyzing the object status information, it can determine the length, width, and height of the object to be moved, thus determining the object's specifications. After obtaining the object specifications, it can calculate the specific position where the robot's end effector, such as the gripper of the robotic arm, should contact and grasp the object when performing the grasping action. This gripping point is crucial for successfully grasping and stably moving the object.

[0068] In some embodiments, step S2 includes S21 to S23, as follows:

[0069] S21, the cloud server determines the object's specifications based on image data from various perspectives, and the object's status information includes image data from each perspective.

[0070] Specifically, the robot's terminal vision sensor can capture images of objects from multiple angles, obtaining image data of the objects from various perspectives. The cloud server can then use advanced image analysis algorithms to perform in-depth processing on this image data, determining the actual length, width, and height of the object, i.e., the object's specifications.

[0071] S22, obtain the maximum arm span of the robot terminal, determine the half perimeter according to the object specifications, and when the maximum arm span is greater than the half perimeter, determine the center point of the opposite side of the object as the gripping point.

[0072] Specifically, after determining the object's dimensions, the cloud server can obtain the key parameter of the robot's maximum arm span from the robot terminal's configuration parameters. The maximum arm span determines the maximum range the robot can reach when grasping the object. Then, it can calculate the object's semi-perimeter based on the object's dimensions. For regular-shaped objects, such as cuboids, the semi-perimeter can be calculated by summing the length and width. When the robot terminal's maximum arm span is greater than the object's semi-perimeter, it means the robot can easily grasp the object around one side. In this case, the cloud server can determine the center points of two opposite sides of the object as the grasping points. For example, when the robot terminal's maximum arm span... When the perimeter is greater than half and the object is a regular cuboid, the four sides of the cuboid, viewed from a normal angle, are the left side, right side, front side, and back side. The left and right sides are opposite sides of the object, as are the front and back sides. Therefore, when the robot's end effector grasps the object, the center points of the left and right sides, or the center points of the front and back sides, can be used as the grasping points. Choosing the center points of opposite sides allows the robot to better balance the weight distribution of the object during grasping, reducing the risk of tilting or falling during handling. The maximum arm span refers to the total length that the two robotic arms of the end effector can reach when fully extended.

[0073] S23, when the maximum arm extension length is less than half the circumference, multi-machine collaboration is triggered to determine the respective transport surfaces of multiple robot terminals, and the point located at the center point gripping offset distance in the transport surface is determined as the gripping point corresponding to each robot terminal.

[0074] Specifically, when the maximum reach is less than half the perimeter, it indicates that a single robot cannot effectively grasp the object independently. In this case, the cloud server triggers a multi-robot collaboration mode, scheduling multiple robot terminals to collaboratively complete the transport task. The cloud server can assign a corresponding transport surface to each participating robot terminal based on the object's specifications and shape. For example, for a cuboid object, four of its six faces can be assigned to four robot terminals, each responsible for a specific face. For each transport surface, the cloud server can calculate its center point and determine a grasping offset distance based on the object's specifications. This offset distance ensures the robot terminal maintains the object's balance during grasping and avoids collisions with other robot terminals. The point on each transport surface, at a distance from the center point of the transport surface from the grasping offset distance, is designated as the grasping point for each robot terminal. Here, multi-robot collaboration refers to multiple robot terminals working together to transport an object; the transport surface refers to the side of the object each robot terminal is responsible for during the collaborative transport process; and the grasping offset distance is the distance offset from the center point of the transport surface when determining the grasping point.

[0075] Based on the above embodiments, step S23 can be implemented in the following ways:

[0076] S231, determine the half-arm length of the robot terminal, and obtain the grasping range corresponding to the half-arm length on the side of each object.

[0077] In multi-robot collaborative handling tasks, understanding the gripping capabilities of the robot end effector is fundamental. The half-arm extension length, which is the length of a single robotic arm, directly limits the range that the robot can reach on the side of an object. Only by first determining the half-arm extension length can we further assess the feasibility of the robot gripping each side of the object, providing a basis for subsequent judgments on whether multi-robot collaboration is needed and how to allocate tasks.

[0078] Specifically, the cloud server can obtain the length data of a single robotic arm from the robot's equipment information database. This data is usually accurately recorded during robot production and configuration and is uploaded to the cloud server for unified management. After obtaining the half-arm extension length, it can combine the object's three-dimensional spatial information and the robot's position and posture around the object to determine the grasping range corresponding to the half-arm extension length on each side of the object through geometric calculations and spatial modeling. For example, drawing a circle with the robot's shoulder joint as the origin and the half-arm extension length as the radius, the area where this circle intersects with the side of the object is the possible grasping range. By determining the half-arm extension length and the corresponding grasping range, the cloud server can clearly understand the capability boundaries of each robot in the current object handling task. This helps to predict in advance whether a single robot can complete the grasping of a certain side of the object, providing accurate data support for subsequent multi-robot collaborative decisions, avoiding situations where robots cannot reach the object during the handling process, and improving the planning efficiency and accuracy of handling tasks.

[0079] Among them, the semi-extended arm length refers to the length of a single robotic arm in the robot terminal, and the grasping range refers to the range that the robotic arm in the robot terminal can cover on the side of each object.

[0080] S232, when the grasping range is smaller than the range threshold of the transport surface, the number of terminals for multi-machine cooperation is determined to be the four-corner preset number, and the sides of adjacent objects are sequentially determined as the transport surfaces corresponding to the respective robot terminals.

[0081] Specifically, the cloud server can compare the gripping range of each handling surface with a pre-set range threshold. This threshold is based on extensive experimentation and practical experience. When a gripping range is found to be smaller than the pre-set threshold, it means a single robot cannot effectively cover that side for gripping, potentially leading to unstable gripping or failure to complete the task. In this case, a pre-set number of robot terminals (typically four) can be used for collaboration. The cloud server automatically determines that four robots are needed and then assigns adjacent object sides to the four robot terminals in a clockwise or counter-clockwise order. This provides stable support and gripping force from the four corners of the object, ensuring balance and stability during transport. Assigning adjacent object sides to the robots is chosen because this arrangement allows for closer and more efficient collaboration between robots, enabling better coordination and joint completion of the transport task. For example, see [link to relevant documentation]. Figure 2 This is a schematic diagram illustrating how to determine the transport surface corresponding to a robot terminal according to an embodiment of the present invention. When the number of terminals in multi-machine collaboration is 4, such as... Figure 2As shown, for the four sides of a regular object, namely object side 1, object side 2, object side 3 opposite to object side 1, and object side 4 opposite to object side 2, object side 1 and object side 2 are adjacent object sides, object side 2 and object side 3 are adjacent object sides, object side 3 and object side 4 are adjacent object sides, and object side 4 and object side 1 are adjacent object sides. Then, object side 1 and object side 2 can be determined as the transport surface corresponding to robot terminal 1, object side 2 and object side 3 can be determined as the transport surface corresponding to robot terminal 2, object side 3 and object side 4 can be determined as the transport surface corresponding to robot terminal 3, and object side 4 and object side 1 can be determined as the transport surface corresponding to robot terminal 4.

[0082] Among them, the range threshold refers to the preset threshold used to measure whether the robot terminal can effectively cover the side of the object and perform grasping operations, and the number of terminals refers to the number of robot terminals when multiple robots cooperate, with a preset number of 4 for the four corners.

[0083] S233, when there is no grasping range smaller than the range threshold of the transport surface, determine the number of terminals for multi-machine cooperation as a preset number diagonally, and determine the diagonally adjacent object sides as the transport surface corresponding to the corresponding robot terminal.

[0084] Specifically, after the cloud server compares the grasping range of all object sides with the range threshold, if it finds no grasping ranges smaller than the threshold, it means that a single robot has sufficient capacity to cover each side for grasping. In this case, a predetermined number of diagonally positioned robots can collaborate. For example, two robots can be used collaboratively. Then, diagonally adjacent object sides can be selected and designated as the handling surfaces for the corresponding robot terminals. This reduces the number of robots used and improves resource utilization efficiency while ensuring handling effectiveness. The reason for assigning diagonally adjacent object sides to the robots is that this layout allows the two robots to form a stable diagonal... Pulling force helps to better control the balance and direction of movement of objects, while avoiding collisions and interference between robots. For example, when the number of terminals in a multi-machine collaboration is 2, the four sides of a regular object are designated as object side 1, object side 2, object side 3, and object side 4. Object side 1 and object side 2 are adjacent, and object side 3 and object side 4 are adjacent. The angle formed by object side 1 and object side 2 is diagonal to the angle formed by object side 1 and object side 2. Therefore, object side 1 and object side 2 can be designated as the handling surface corresponding to robot terminal 1, and object side 3 and object side 4 can be designated as the handling surface corresponding to robot terminal 2. The preset number of diagonals is 2.

[0085] S234, determine the point on the transport surface located at the center point gripping offset distance as the gripping point corresponding to each robot terminal.

[0086] Specifically, after determining the transport surface corresponding to each robot terminal, the offset direction corresponding to each robot terminal can be determined on the transport surface. On the offset direction corresponding to each robot terminal, the point located at the center point gripping offset distance is determined as the gripping point corresponding to the robot terminal. When determining the offset direction corresponding to each robot terminal, it can be determined according to the transport surface corresponding to each robot terminal. For example, when the transport surface corresponding to robot terminal 1 is object side 1 and object side 2, there is a common edge between object side 1 and object side 2. The midpoint of this common edge can be obtained, and the direction from the center point of object side 1 to the midpoint of the common edge can be obtained. This direction can be determined as the offset direction of the robot terminal on object side 1. Similarly, the direction from the center point of object side 2 to the midpoint of the common edge can be determined as the offset direction of object side 2.

[0087] In some embodiments, the gripping offset distance may be preset. In other embodiments, the gripping offset distance may be determined by the following steps:

[0088] S2341, calculate the object volume based on the object specifications, and obtain the offset coefficient based on the ratio of the reference volume to the object volume.

[0089] Specifically, based on the object's specifications, the corresponding volume of the object can be calculated. For example, for a regular object, after obtaining the object's length, width, and height, the object's volume can be obtained by multiplying the length, width, and height. The corresponding offset coefficient can be obtained by calculating the ratio between the reference volume and the object's volume.

[0090] Among them, the object volume refers to the volume calculated according to the object's specifications, the reference volume refers to the preset reference volume value used for comparison with the object volume, and the offset coefficient is the ratio between the reference volume and the object volume, which can be used to indicate the degree of deviation of the object volume from the reference volume.

[0091] S2342, the grabbing offset distance is obtained by multiplying the offset coefficient and the reference offset distance and then performing weighted processing.

[0092] Specifically, the corresponding grabbing offset distance can be obtained by multiplying the offset coefficient by a preset baseline offset distance and then weighting the resulting value. The baseline offset distance refers to the offset distance under preset standard conditions.

[0093] The above implementation method can ensure the stability and balance of the robot terminal when grasping objects, and prevent objects from tilting or falling during transportation.

[0094] S3: The cloud server determines the constraint section based on environmental data, calculates the trajectory offset value based on the object specifications and section parameters, and generates control commands based on the gripping point and trajectory offset value, which are then sent to the robot terminal.

[0095] Specifically, after obtaining the object's specifications, during the object handling process, the cloud server can perform a comprehensive and detailed analysis of the surrounding environment based on the environmental data uploaded by the robot terminal in real time. This allows for the accurate identification of constraint sections, which encompass various areas requiring careful handling by the robot, such as sections with obstacles ahead or sections requiring turns. The turning angles of these sections will affect the robot's movement. When a constraint section is identified, indicating a need for a turn or an obstacle ahead, the cloud server can begin trajectory offset planning. In this process, object specifications are a crucial consideration. Different object sizes may affect the robot's flexibility and stability during handling. For example, larger objects... Turning requires a larger turning angle to avoid collisions with surrounding obstacles. A smoother maneuver is also necessary to prevent objects from falling or the robot from becoming unbalanced due to inertia. Meanwhile, path parameters play a crucial role in trajectory offset planning. These parameters primarily refer to the turning angle at corners. Different turning angles require the robot to perform turning operations at different speeds and angles. The cloud server can integrate object specifications and path parameters, employing advanced path planning algorithms to perform complex and precise calculations to determine the trajectory offset value. This trajectory offset value specifies in detail the direction and angle of deviation the robot needs to make from its original trajectory when facing constrained path segments, thus enabling the planning of a safe and efficient transport path.

[0096] After obtaining the trajectory offset value, the cloud server integrates it with the previously calculated gripping point information. The gripping point determines how the robot stably grips the object, while the trajectory offset value determines the robot's path during the handling process. By combining these two, the cloud server can generate a set of detailed and precise control instructions. Furthermore, the cloud server can quickly and accurately send these control instructions to the robot terminal through the communication module, guiding the robot to complete the object handling task efficiently and safely in complex working environments.

[0097] Among them, the constrained section refers to the section of the road that the robot needs to handle with caution due to the presence of obstacles or the need to turn during the process of the robot carrying objects. The section parameter refers to the turning angle at the corner. The trajectory offset value refers to the direction and angle of the robot's deviation from the original trajectory when facing the constrained section. The control command refers to a set of detailed and precise instructions generated by the cloud server based on the gripping point information and the trajectory offset value to guide the robot's operation.

[0098] In some embodiments, step S3, "the cloud server determines the constrained road segment based on environmental data and calculates the trajectory offset value based on object specifications and road segment parameters," includes the following steps:

[0099] S31, the cloud server identifies passable and impassable areas based on the pixel classification results in the environmental data, wherein the environmental data includes image data corresponding to the frontal view.

[0100] Specifically, when the sensor module of the robot terminal collects environmental data, it can acquire frontal view image data. The cloud server can preprocess the frontal view image data in the uploaded environmental data, such as using filtering algorithms to remove noise interference in the image. After preprocessing, the preprocessed image data can be input into a pre-trained semantic segmentation model for inference. This model is trained on a large amount of image data, which covers various different working environment scenarios, including warehouses with different layouts, various obstacle types, and complex passage situations. Through deep learning algorithms, the model learns the feature patterns of different semantic categories in the image. When the input image enters the model, the model analyzes and judges each pixel in the image and outputs the classification result of each pixel, that is, it clarifies which semantic category each pixel belongs to. For example, the model can identify whether a pixel belongs to the ground, passage, obstacle, or wall.

[0101] Based on the pixel classification results output by the semantic segmentation model, the cloud server can further label the image regions. For areas such as the ground and passageways that the robot can safely pass through, they are marked as passable areas. These areas provide a safe path for the robot to move freely and perform transport tasks. For areas such as obstacles and walls that will block the robot's progress, the cloud server marks them as impassable areas. These areas clearly define the range that the robot needs to avoid during its movement, preventing collisions with obstacles and ensuring the safe conduct of transport operations. Through this precise region labeling, the cloud server provides a clear and accurate environmental information foundation for subsequently determining constraint segments and planning the robot's trajectory, enabling the robot to complete transport tasks safely and efficiently in complex working environments.

[0102] Among them, the pixel classification result refers to the category to which each pixel belongs after analyzing and judging each pixel in the image. The passable area refers to the area that the robot can safely pass through, such as the ground or passage, according to the pixel classification result. The impassable area refers to the area that the robot will be blocked from moving forward, such as obstacles or walls, according to the pixel classification result.

[0103] S32, obtain the path centerline of the passable area. When the angle between the path centerline and the current driving direction of the robot terminal is greater than the threshold, retrieve the area map.

[0104] After identifying the passable area, the cloud server can obtain the path centerline of the passable area through a skeleton extraction algorithm. The path centerline represents the ideal driving path of the robot within the passable area. The cloud server can compare the path centerline with the current driving direction of the robot terminal. When the angle between the path centerline and the current driving direction of the robot terminal exceeds a set threshold, it clearly indicates that if the robot continues to drive in the current direction, it will not be able to move along the ideal path centerline and needs to turn. Once it is determined that the robot needs to turn, the cloud server will quickly retrieve the area map. The area map contains comprehensive information about the working environment, with detailed distribution of passable and impassable areas. By retrieving the area map, the cloud server can obtain richer information, providing sufficient data support for more accurately determining the constraint segments and planning the robot's turning path. This ensures that the robot can smoothly adjust its direction of travel in complex environments and complete the handling task efficiently and safely.

[0105] Among them, the path centerline refers to the ideal driving path of the robot terminal, the current driving direction refers to the current driving direction of the robot terminal, the threshold refers to the pre-set angle value used to determine whether the angle between the path centerline and the current driving direction of the robot is too large, and the area map refers to the map containing the working environment of the area where the robot terminal is located.

[0106] S33, determine the location point of the robot terminal in the area map, and determine the road segment where the location point is located and the road segment that the robot terminal has not passed through as the constraint road segment.

[0107] Specifically, the cloud server can use the robot terminal's positioning system information to determine its location point on the regional map. The location point can accurately reflect the robot terminal's current position in the entire working environment. Then, the cloud server uses the location point as a clue to find the road segment where the location point is located but which the robot terminal has not yet passed. These road segments can be identified as constraint road segments. Constraint road segments are usually areas that the robot needs to pay close attention to during subsequent travel, which may be due to obstacles ahead or complex road segments that require turning.

[0108] After the cloud server determines the constrained road segments based on environmental data, this solution also includes the following embodiments:

[0109] A1: Obtain the width of the target item in the direction perpendicular to the current driving direction, and obtain the width of the constrained road segment.

[0110] Specifically, after determining the constrained road segment, when the robot terminal is carrying the target item through the constrained road segment, it needs to determine whether the robot terminal can successfully pass through the constrained road segment. At this time, the width of the target item in the direction perpendicular to the current travel direction can be obtained, and the cloud server can obtain the width of the constrained road segment from the regional map. By comparing the width of the target item in the direction perpendicular to the current travel direction with the width of the road segment, it is possible to intuitively determine whether the robot terminal can successfully pass through the constrained road segment with the current carrying posture. If the width of the item is less than the width of the road segment, the robot terminal can usually pass through smoothly. If the width of the item is greater than the width of the road segment, it may not be able to pass through smoothly, and the angle of the target item may need to be adjusted accordingly. Here, the target item refers to the item carried by the robot terminal, the item width refers to the width of the target item in the direction perpendicular to the current travel direction, and the road segment width refers to the width of the constrained road segment.

[0111] A2. When the width of the item is greater than or equal to the width of the road segment, a reference model corresponding to the item specifications is generated. The reference model is adjusted according to the preset rotation angle, and the reference width corresponding to the reference model is obtained.

[0112] Specifically, when the width of an item is greater than or equal to the width of the road segment, the robot terminal may not be able to smoothly transport the item through the constrained road segment. When the target item is angle-adjustable, such as a wooden product, the robot terminal can adjust the angle of the target item it is transporting. If the target item is not angle-adjustable, such as a glass product, a new transport path can be used. When adjusting the angle of an angle-adjustable target item, the robot terminal can place the target item on the ground, and then first construct a reference model corresponding to the dimensions of the target item, and initially... The reference model is positioned identically to the target item. After constructing the reference model, multiple rotation angles can be preset, such as 45° and 90°. The reference model is adjusted according to the preset rotation angle and direction. Each adjustment yields the width of the reference model in the direction perpendicular to the target item's current direction of travel, i.e., the reference width. For example, if the preset rotation direction is clockwise, the reference model can be rotated 45 degrees clockwise for the first adjustment, yielding the reference width in the direction perpendicular to the target item's current direction of travel. Here, the reference model refers to the virtual model corresponding to the target item, the preset rotation angle is the pre-defined angle at which the reference model is rotated, and the reference width is the width of the reference model in the direction perpendicular to the target item's current direction of travel at the corresponding preset rotation angle.

[0113] A3, determine the rotation angle where the reference width is less than the road segment width as the target angle, and send the target angle to the robot terminal to adjust the angle of the item.

[0114] Specifically, when the reference width is less than the road segment width, the rotation angle corresponding to the current reference width of the reference model is determined as the target angle. For example, when the reference model is rotated 90° clockwise, its corresponding reference width is less than the road segment width, so 90° can be determined as the target angle. After determining the target angle, the cloud server can send the target angle information to the robot terminal in a specific data format through a reliable communication link established with the robot terminal. After receiving the target angle information, the robot terminal can make corresponding angle adjustments to the target item. By adjusting the angle of the target item, the robot terminal can transport the item through the constrained road segment at a suitable angle, ensuring the safe and efficient completion of the transport task.

[0115] If, after rotating the reference model one full turn, the corresponding reference width is consistently greater than the road segment width, then it can be assumed that the target item may be unable to pass through the constrained road segment, and a new path can be taken. Here, the target angle refers to the rotation angle of the reference model when the reference width is less than the road segment width.

[0116] S34, obtain the turning angle of the constrained road segment, calculate the turning speed of the robot terminal from the current driving direction to the direction corresponding to the turning angle according to the item specifications, the road segment parameters include the turning angle, and the trajectory offset value includes the turning speed.

[0117] For a given constrained road segment, the cloud server first needs to obtain its turning angle. This turning angle is one of the key parameters for calculating the trajectory offset. The turning angle can be obtained by reading road segment information from the regional map or by analyzing the geometry of the constrained road segment using image recognition technology. Then, based on the previously determined object specifications, the cloud server can calculate the turning speed of the robot terminal from the current driving direction to the direction corresponding to the turning angle. Larger objects require slower speeds when turning to ensure sufficient time and space to complete the turn and avoid collisions with surrounding obstacles, while lighter and smaller objects can have their turning speeds increased appropriately. By comprehensively considering the object specifications and turning angle, and using specific physical models and algorithms, a suitable turning speed can be calculated.

[0118] Among them, the turning angle refers to the angle that the robot terminal needs to change from the current straight-line driving direction to the target turning direction during the driving process, and the turning speed refers to the speed of the robot terminal during the turning process.

[0119] In some embodiments, "obtaining the turning angle of the constrained road segment" in step S34 includes the following steps:

[0120] S341, Obtain the movement trajectory of the robot terminal in the area map within a preset time period, and determine the tangent direction of the end point of the movement trajectory as the driving direction of the robot terminal.

[0121] Specifically, the cloud server can extract the robot terminal's movement trajectory information within a pre-set time period from the regional map. For example, it can obtain the movement trajectory within the past 10 seconds. After obtaining the robot terminal's movement trajectory, the cloud server can analyze the trajectory curve at the trajectory endpoint to determine the tangent direction corresponding to the trajectory endpoint, and determine the robot terminal's driving direction as the tangent direction.

[0122] Among them, the preset time period refers to a pre-set time period, the movement trajectory refers to the trajectory of the robot terminal within the preset time period, and the driving direction refers to the direction corresponding to the movement of the robot terminal.

[0123] S342, determine the centerline of the constrained road segment. When the tangent directions of all path points on the centerline are consistent, determine the angle between the centerline and the driving direction as the turning angle, and determine the direction from the driving direction to the centerline as the turning direction.

[0124] Specifically, after determining the robot terminal's travel direction, the cloud server can obtain the centerline of the constrained road segment using image processing techniques such as skeleton extraction algorithms. For constrained road segments, whether they are straight passages, curves requiring turning, or sections with obstacles that need to be detoured, the centerline represents the most suitable path for the robot terminal. When analyzing the centerline, if the tangent directions of all path points on the centerline are consistent, the constrained road segment can be considered relatively regular, possibly a straight passage. In this case, the cloud server can use geometric calculation methods such as trigonometric functions to calculate the angle between the centerline and the robot terminal's travel direction. This angle is the turning angle that the robot terminal needs to turn when passing through the corresponding constrained road segment, i.e., the turning angle. Furthermore, by comparing the spatial relationship between the centerline and the travel direction, the direction from the travel direction to the centerline can be determined as the robot terminal's turning direction.

[0125] See Figure 3 This is a schematic diagram of determining the turning direction provided by an embodiment of the present invention, such as... Figure 3 As shown, based on the tangent direction of the robot terminal's trajectory endpoint within a preset time period, the robot terminal's travel direction can be determined, and the centerline of the constrained road segment can be obtained. Figure 3 As can be seen, there is an angle between the driving direction and the centerline of the constrained road section. This angle can be determined as the turning angle, and the direction from the driving direction to the centerline can be determined as the turning direction.

[0126] The centerline refers to a representative path of the constrained road segment, representing the most suitable path for the robot terminal to traverse. The turning direction refers to the direction the robot terminal needs to take when turning from the current travel direction to the centerline direction.

[0127] S343, when the tangent directions of the path points on the center line are inconsistent, the path point on the center line that is closest to the positioning point of the robot terminal is determined as the starting point.

[0128] Specifically, when analyzing the centerline of a constrained road segment, if inconsistent tangent directions are found at path points, it can be assumed that there may be multiple path segments requiring turns within that constrained road segment. In this case, determining a suitable starting point is crucial for accurately calculating the turning angle and direction. The cloud server can obtain the robot terminal's location point on the area map. This location point is fed back to the cloud server in real time via the robot terminal's built-in positioning system, such as GPS. Then, the cloud server can search for the nearest path point to this location point on the centerline of the constrained road segment and determine it as the starting point. Selecting the nearest path point as the starting point can best match the robot's current actual position and movement trend, making subsequent calculations of turning angles and directions more accurate and effective. The starting point refers to the path point on the centerline closest to the robot terminal.

[0129] S344: Starting from the starting point, obtain the path segments composed of path points with the same tangent direction as the starting point, determine the angle between the path segment and the driving direction as the turning angle, and determine the direction from the driving direction to the path segment as the turning direction.

[0130] After determining the starting point, the cloud server uses this starting point as the traversal origin and searches for path points along the centerline of the constrained road segment. Using a curve tangent direction consistency judgment method, it checks each path point along the centerline, filtering out those with the same tangent direction as the starting point. These path points collectively form a path segment, representing a path with the same turning tendency starting from the robot's current position. After obtaining the path segment, the cloud server calculates the angle between the centerline of the path segment and the robot's terminal travel direction, determining this angle as the turning angle, and identifying the direction from the travel direction to the path segment as the turning direction. Here, the traversal origin refers to the starting point for traversing path points with the same tangent direction, and a path segment refers to a path segment on the constrained road segment with the same turning tendency.

[0131] In some embodiments, step S34, "calculating the turning speed of the robot terminal from the current driving direction to the direction corresponding to the turning angle based on the item specifications," includes the following steps:

[0132] S345, calculate the volume of an object based on its specifications, and obtain the first adjustment coefficient by weighting the reciprocal of the object's volume.

[0133] Specifically, based on the object's specifications, its volume can be calculated. For example, if the object's specifications are its length, width, and height, the volume can be obtained by multiplying these dimensions. After obtaining the volume, the reciprocal of the volume is weighted to obtain a first adjustment coefficient. This weighting process considers various factors, such as the stability requirements of different types of objects during turning and the robot's load-bearing capacity. Larger objects have smaller reciprocals of their volumes, resulting in a smaller first adjustment coefficient after weighting. This means that the turning speed will be reduced in subsequent calculations to accommodate the need for more time and space to complete the turn and avoid collisions with surrounding obstacles. The first adjustment coefficient refers to the coefficient used to adjust the speed based on the object's volume.

[0134] S346, Obtain item attributes. When the item attribute is fragile, obtain the preset coefficient corresponding to the risk level of the target item as the second adjustment coefficient.

[0135] Specifically, the cloud server can obtain the attribute information of the items being handled by the robot terminal. Item attributes cover important characteristics such as the item's material and whether it is fragile. When an item is classified as fragile, the cloud server can further obtain a preset coefficient corresponding to the risk level of the target item as a second adjustment coefficient. The risk level of a target item is usually classified according to factors such as the fragility of its material and its value. For example, glass products may have a higher risk level, while plastic products may have a relatively lower risk level. Each risk level has a corresponding preset coefficient, which is derived from a large number of experiments and practical experience. For example, assuming that the risk levels are divided into three levels—high, medium, and low—with corresponding preset coefficients of 0.5, 0.7, and 0.9, respectively, when the target item is determined to be a high-risk fragile item, the corresponding second adjustment coefficient is 0.5. This coefficient is used to further reduce the turning speed when calculating the turning speed to ensure the safety of the fragile item during the turning process.

[0136] Among them, item attributes refer to the inherent characteristics of the item itself, fragility attributes refer to the characteristics of the item that are easily damaged when subjected to external forces, target item refers to the item that the robot terminal is handling, risk level refers to the risk classification of the target item based on its fragility, which can be used to assess the risk of loss that the item may suffer during handling, preset coefficient refers to the value set in advance for each risk level, which is used to adjust when calculating turning speed, and second adjustment coefficient refers to the coefficient for adjusting speed determined based on item attributes.

[0137] S347, the total adjustment coefficient is obtained by comprehensively calculating the first adjustment coefficient and the second adjustment coefficient, and the turning speed is obtained by multiplying the total adjustment coefficient and the reference speed.

[0138] Specifically, after obtaining the first and second adjustment coefficients, the cloud server performs a comprehensive calculation to obtain the total adjustment coefficient. The calculation method for the total adjustment coefficient can be selected according to the actual situation. For example, the corresponding weighting factors can be determined according to the importance attached to the object volume and the item's fragility attribute. The first and second adjustment coefficients are then simply weighted and summed according to the weighting factors corresponding to the object volume and the item's fragility attribute to obtain the corresponding total adjustment coefficient. Then, the turning speed can be obtained by multiplying the total adjustment coefficient by the pre-set baseline speed. Through this calculation method, factors such as the item's specifications and turning angle are comprehensively considered to determine a suitable speed for the robot's turning operation in constrained road sections.

[0139] The total adjustment coefficient refers to the coefficient obtained by comprehensively calculating the first and second adjustment coefficients, and the reference speed refers to the robot's preset driving speed under ideal conditions, without considering the size of the object or the road conditions.

[0140] The above-described implementation methods can effectively reduce the risk of objects falling or being damaged during transportation, while also improving the efficiency of robot terminals in passing through constrained road sections.

[0141] S4 acquires real-time status data of the robot terminal when executing control commands, and the cloud server generates motion adjustment information based on the real-time status data to update the robot terminal's commands.

[0142] After receiving control commands from the cloud server, the robot terminal can perform corresponding turning operations during object handling. During this process, its sensor modules continuously collect real-time status data, which is then constantly transmitted to the cloud server. This real-time status data includes vibration data, a crucial indicator of robot stability. Low vibration values ​​indicate smooth robot operation and stable object handling, while increased vibration values ​​may indicate encountering instability, such as uneven ground or excessive speed, which can cause the object to sway during handling. This can affect the successful completion of tasks. The cloud server can analyze the received real-time status data, especially vibration data. When abnormal vibration data is detected, the cloud server can generate corresponding adjustment information, namely motion adjustment information. The motion adjustment information can adjust the speed of the robot terminal to reduce or eliminate vibration, thereby maintaining the stability of the object during the handling process. After receiving this motion adjustment information, the robot terminal will immediately execute the corresponding instruction update and adjust its own movement speed. Through this real-time status monitoring and instruction update mechanism, the cloud server can ensure that the robot always maintains the best state when executing control instructions, and can make rapid adjustments even when encountering unstable factors, thereby greatly improving the stability of object handling.

[0143] Among them, real-time status data refers to the data that the robot terminal continuously collects and sends to the cloud server through its sensor module during the execution of control commands, and motion adjustment information refers to the instruction information generated by the cloud server after analyzing the received real-time status data, which is used to adjust the motion state of the robot terminal.

[0144] Based on the above embodiments, step S4 can be implemented in the following ways:

[0145] S41, the sensor module based on the robot terminal acquires vibration values, and the real-time status data includes vibration values.

[0146] Specifically, the sensor modules equipped on the robot terminal can detect vibrations generated during the handling process due to various reasons such as uneven ground or excessive speed, and convert them into quantifiable values, i.e., vibration values. These vibration values ​​can be continuously sent to the cloud server for analysis as part of the real-time status data. The higher the vibration value, the more unstable the robot terminal is during the handling process, and the more likely it is that the object may fall. The vibration value refers to the quantified value of the robot's vibration intensity collected by the robot terminal's vibration sensors.

[0147] S42, when the vibration value is greater than the vibration threshold, obtain the vibration difference between the vibration value and the reference vibration value, and calculate the speed adjustment value by weighting the ratio of the vibration difference to the reference difference.

[0148] Specifically, after receiving real-time status data from the robot terminal, the cloud server first checks whether the vibration value exceeds a preset vibration threshold. The vibration threshold is a preset limit used to determine if the robot terminal's current vibration level is within an acceptable range. If the vibration value exceeds the threshold, the robot terminal may be unstable during object handling and requires adjustment. At this point, the cloud server calculates the difference between the current vibration value and a baseline vibration value, known as the vibration difference. The baseline vibration value is a preset value representing the ideal vibration level of the robot terminal. The vibration difference reflects the deviation between the current vibration state and the ideal state. The vibration difference is then calculated by comparing the vibration difference with the baseline difference and performing a weighted average to obtain a speed adjustment value. Based on this speed adjustment value, the robot terminal's speed can be adjusted accordingly. Here, the vibration threshold is a preset limit value used to determine if the vibration level exceeds the standard; the baseline vibration value is a preset value representing the ideal vibration state; the vibration difference is the difference between the current vibration value and the baseline vibration value; the baseline difference is the vibration difference under a preset standard state; and the speed adjustment value is the value used to adjust the robot terminal's speed.

[0149] S43: Obtain motion adjustment information by subtracting the speed adjustment value from the current speed of the robot terminal, and update the speed of the robot terminal based on the motion adjustment information.

[0150] Specifically, the cloud server can obtain the current speed of the robot terminal, subtract the speed adjustment value from the current speed to obtain the motion adjustment information, and update the speed of the robot terminal accordingly based on the motion adjustment information.

[0151] The above implementation methods can improve the stability of the robot terminal when handling objects.

[0152] See Figure 4 This is a schematic diagram of the structure of a cloud-based robot control system provided in an embodiment of the present invention. The data processing system of this cloud-based robot control system includes:

[0153] The upload module is used by the robot terminal to acquire environmental data and object status information based on the sensor module, and to upload the environmental data and object status information to the cloud server based on the communication module.

[0154] The determination module is used by the cloud server to determine the object's specifications based on the object's state information and to calculate the grab points based on the object's specifications.

[0155] The calculation module is used by the cloud server to determine the constraint road segment based on environmental data, calculate the trajectory offset value based on the object size and road segment parameters, and generate control commands based on the gripping point and trajectory offset value and send them to the robot terminal.

[0156] The generation module is used to acquire real-time status data of the robot terminal when executing control commands. The cloud server generates action adjustment information based on the real-time status data to update the robot terminal's commands.

[0157] Figure 4 The apparatus of the illustrated embodiment can be used to perform corresponding actions. Figure 1 The steps in the method embodiments shown are implemented in a similar manner and have similar technical effects, and will not be repeated here.

[0158] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for controlling a robot via the cloud, characterized in that, include: The robot terminal acquires environmental data and object status information based on the sensor module, and uploads the environmental data and object status information to the cloud server based on the communication module; The cloud server determines the object's dimensions based on its state information and calculates the grab points based on those dimensions, including: The cloud server determines the object's dimensions based on image data from various perspectives, and the object's status information includes image data from each perspective. Obtain the maximum reach of the robot terminal, determine the half-circumference based on the object's specifications, and when the maximum reach is greater than the half-circumference, determine the center point of the opposite side of the object as the gripping point. When the maximum reach length is less than half the circumference, multi-robot collaboration is triggered. The respective transport surfaces of multiple robot terminals are determined, and the points located at the center point gripping offset distance within these transport surfaces are identified as the gripping points for each robot terminal, including: Determine the half-arm length of the robot terminal and obtain the grasping range corresponding to the half-arm length on the side of each object; When the grasping range is smaller than the range threshold of the transport surface, the number of terminals for multi-machine collaboration is determined to be the preset number at the four corners, and the sides of adjacent objects are sequentially determined as the transport surfaces corresponding to the respective robot terminals. When there is no grasping range smaller than the transport surface range threshold, the number of terminals for multi-machine collaboration is determined to be a pre-set number diagonally, and the diagonally adjacent object sides are determined to be the transport surface corresponding to the corresponding robot terminal. The point located at the center point of the handling surface at the gripping offset distance is determined as the gripping point corresponding to each robot terminal; The cloud server determines the constrained road segment based on environmental data, calculates the trajectory offset value based on the object specifications and road segment parameters, and generates control commands based on the gripping point and trajectory offset value, which are then sent to the robot terminal. The system acquires real-time status data of the robot terminal when executing control commands, and the cloud server generates action adjustment information based on the real-time status data to update the robot terminal's commands.

2. The method according to claim 1, characterized in that, The grab offset distance is determined by the following steps: The volume of an object is calculated based on its specifications, and an offset coefficient is obtained based on the ratio of the reference volume to the object's volume. The capture offset distance is obtained by multiplying the offset coefficient and the reference offset distance and then performing a weighted summation.

3. The method according to claim 1, characterized in that, The cloud server determines the constrained road segments based on environmental data and calculates the trajectory offset values ​​based on object specifications and road segment parameters, including: The cloud server identifies passable and impassable areas based on pixel classification results in environmental data, which includes image data corresponding to the frontal view. Obtain the centerline of the passable area. When the angle between the centerline of the path and the current travel direction of the robot terminal is greater than a threshold, retrieve the area map. Determine the location of the robot terminal in the area map, and identify the road segments where the location point is located and which the robot terminal has not yet traversed as constraint road segments; Obtain the turning angle of the constrained road segment, and calculate the turning speed of the robot terminal from the current driving direction to the direction corresponding to the turning angle based on the specifications of the item. The road segment parameters include the turning angle, and the trajectory offset value includes the turning speed.

4. The method according to claim 3, characterized in that, Obtain the turning angle of the constrained road segment, including: Obtain the movement trajectory of the robot terminal in the area map within a preset time period, and determine the tangent direction of the end point of the movement trajectory as the driving direction of the robot terminal; Determine the centerline of the constrained road segment. When the tangent directions of all path points on the centerline are consistent, determine the angle between the centerline and the driving direction as the turning angle, and determine the direction from the driving direction to the centerline as the turning direction. When the tangent directions of the path points on the center line are inconsistent, the path point on the center line that is closest to the positioning point of the robot terminal is determined as the starting point. Starting from the starting point, obtain path segments composed of path points with the same tangent direction as the starting point. Determine the angle between the path segment and the driving direction as the turning angle, and determine the direction from the driving direction to the path segment as the turning direction.

5. The method according to claim 4, characterized in that, Calculate the turning speed of the robot terminal from the current driving direction to the direction corresponding to the turning angle based on the specifications of the item, including: The volume of an object is calculated based on its specifications, and the first adjustment coefficient is obtained by weighting the reciprocal of the object's volume. Obtain the item's attributes. When the item's attribute is fragile, obtain the preset coefficient corresponding to the risk level of the target item as the second adjustment coefficient. The total adjustment coefficient is obtained by combining the first and second adjustment coefficients. The turning speed is obtained by multiplying the total adjustment coefficient and the reference speed.

6. The method according to claim 3, characterized in that, After the cloud server determines the constrained road segments based on environmental data, it also includes: Obtain the width of the target item in the direction perpendicular to the current driving direction, and obtain the width of the constrained road segment; When the width of an item is greater than or equal to the width of a road segment, a reference model corresponding to the item's specifications is generated. The reference model is then adjusted according to a preset rotation angle, and the reference width corresponding to the reference model is obtained. The target angle is determined when the reference width is smaller than the road segment width. The target angle is then sent to the robot terminal to adjust the angle of the item.

7. The method according to claim 1, characterized in that, The system acquires real-time status data of the robot terminal while executing control commands. Based on this real-time status data, the cloud server generates motion adjustment information to update the robot terminal's commands, including: Vibration values ​​are acquired by the sensor module of the robot terminal, and the real-time status data includes vibration values. When the vibration value is greater than the vibration threshold, the vibration difference between the vibration value and the reference vibration value is obtained, and the speed adjustment value is obtained by weighted calculation of the ratio of the vibration difference to the reference difference. The motion adjustment information is obtained by subtracting the speed adjustment value from the current speed of the robot terminal, and the speed of the robot terminal is updated based on the motion adjustment information.

8. A system for cloud-controlled robots implementing the method of claim 1, characterized in that, include: The upload module is used by the robot terminal to acquire environmental data and object status information based on the sensor module, and to upload the environmental data and object status information to the cloud server based on the communication module. The determination module is used by the cloud server to determine the object's specifications based on the object's state information and to calculate the grab points based on the object's specifications. The calculation module is used by the cloud server to determine the constraint road segment based on environmental data, calculate the trajectory offset value based on the object size and road segment parameters, and generate control commands based on the gripping point and trajectory offset value and send them to the robot terminal. The generation module is used to acquire real-time status data of the robot terminal when executing control commands. The cloud server generates action adjustment information based on the real-time status data to update the robot terminal's commands.

Images