Mechanical arm motion trail simulation verification method based on virtual scene
By recording the robotic arm operating trajectory in a virtual scene and performing similarity and collision assessment, the equipment limitation and inaccurate evaluation of traditional robotic arm training is solved, and efficient and diverse training and evaluation solutions are provided, improving the skills and team level of operators.
Patent Information
- Application Number
- CN202510613341.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-13
- Publication Date
- 2025-08-29
AI Technical Summary
The existing robotic arm operation training methods lack intuitive experience, traditional practical operations are limited by equipment and venues, and the evaluation results lack objectivity and unity, making it difficult to provide diversified and efficient training and evaluation.
By constructing a virtual scene, recording the robotic arm operation trajectory, calculating the trajectory similarity and collision evaluation coefficients, objective evaluation is performed using computer algorithms, combining the diversified training environment of the virtual scene and personalized training plan.
It has achieved efficient and diversified training for robotic arm operators, improved the objectivity and accuracy of adaptability and evaluation, reduced training costs and time, and improved the operating skills level.
Smart Images
Figure CN120564520A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of simulation verification, and in particular to a method for simulating and verifying a robot arm's motion trajectory based on a virtual scene. Background Art
[0002] Robotic arms are increasingly used in modern industrial automation. Their precise and efficient operation significantly improves production efficiency and product quality. However, fully utilizing the robot's capabilities requires the ability to set a specific route. This requires operators to not only be familiar with the basic operation of the robot but also master the complex skills of designing routes. Route design involves precisely planning the robot's trajectory, including every detail of the starting point, end point, and intermediate paths. This requires operators to possess solid mechanical engineering knowledge, computer programming skills, and a deep understanding of three-dimensional space. Therefore, systematic operator training is crucial. Through training, operators can learn how to flexibly design optimal movement routes for the robot based on different task requirements, thereby ensuring smooth production processes. Furthermore, a comprehensive evaluation system is essential to accurately assess each operator's mastery of this skill, allowing for timely identification and improvement of deficiencies. This will continuously improve the overall team's operational capabilities and provide a solid technical foundation for efficient production.
[0003] Currently, training and testing for robotic arm operators primarily relies on practical exercises based on real-world scenarios or theoretical instruction. While these methods can help operators understand and master relevant skills to a certain extent, they also have significant limitations. First, practical exercises in real-world scenarios are often limited by equipment, space, and time constraints, making it difficult to provide diverse training scenarios and thus restricting operators' ability to handle complex tasks. Second, while theoretical instruction can provide a systematic knowledge framework, it lacks the intuitive experience of actual operation, making it difficult to effectively integrate theory with practice. Furthermore, the current reliance on manual judgment to demonstrate training effectiveness is inefficient and prone to lack of standardized and objective scoring. Different evaluators may provide different evaluations due to subjective factors, which not only affects the accuracy of the evaluation but also makes it difficult to accurately identify and effectively improve the operator's skill level. Therefore, a more efficient, objective, and diversified training and assessment method is urgently needed to address the shortcomings of existing methods, improve operator training effectiveness and skill levels, and better meet the needs of modern industrial automation.
[0004] In view of this, a robot arm motion trajectory simulation verification method based on virtual scenes is needed. Summary of the Invention
[0005] In response to the problems in the existing technology of lacking intuitive experience of actual operation, and the corresponding training and evaluation methods being time-consuming and not objective enough, the present invention provides a method for simulating and verifying the motion trajectory of a robotic arm based on a virtual scene. The method can build a virtual scene, then obtain the motion path of the robotic arm formed by the operator operating the robotic arm in the virtual scene, compare the motion path with the standard path, and finally obtain an evaluation coefficient based on the similarity and whether it collides with obstacles. Based on the evaluation coefficient, it can be seen whether the operator's operation is standard or whether it has a high score. The specific technical solution is as follows:
[0006] A method for simulating and verifying the motion trajectory of a robotic arm based on a virtual scene comprises the following steps:
[0007] Constructing a virtual scene, wherein the virtual scene at least includes a robotic arm, a target object, a starting point for the object's movement, and an end point for the object's movement, wherein the robotic arm is provided with a gripping end point;
[0008] Record the coordinates of the gripping endpoint of the robotic arm at each time point (x i ,y i ,z i ), according to the time sequence, we get the first motion trajectory T1 = {(x1, y1, z1), (x2, y2, z2), …, (x n ,y n ,z n ),};
[0009] defining a standard motion trajectory and calculating its similarity with the first motion trajectory;
[0010] Calculate the trajectory evaluation coefficient as follows:
[0011]
[0012] Where F(S) is the similarity between the first motion trajectory and the standard motion trajectory, O i is the collision coefficient of the first motion trajectory at time point i, β i is the collision weight coefficient at time point i, α s , α o are the weight coefficients of similarity and collision coefficient respectively.
[0013] Preferably, the calculation of the evaluation coefficient also takes into account the degree of inclination of the robotic arm when picking up an object. The specific calculation formula is as follows:
[0014]
[0015] Among them, F(RS) i Indicates the tilt coefficient, which is used to reflect the standard degree of tilt at the clamping endpoint time point i, μ irepresents the tilt weight coefficient at time point i, α RS Indicates the weight coefficient of the tilt coefficient.
[0016] Preferably, the tilt coefficient F(RS) is represented by a similarity compared with standard angle data.
[0017] Preferably, the process of obtaining the tilt coefficient F(RS) is as follows:
[0018] Record the rotation angle coordinates of the gripping end point of the robot arm at each time point (Rx i ,Ry i ,Rz i ), and obtaining first rotation time series data of the rotation of the target object in chronological order;
[0019] defining standard rotation time series data for verifying whether the first rotation time series data meets the standard;
[0020] The similarity between the first rotation time series data and the standard rotation time series data is calculated as the tilt coefficient F(RS).
[0021] Preferably, for each target object type, at least one corresponding standard rotation condition is set.
[0022] Preferably, the weight coefficient α of similarity and collision coefficient s , α o The value of is 1.
[0023] Preferably, the operator operates the simulated robotic arm and moves the endpoint of the robotic arm by clicking the mouse, thereby setting the movement trajectory of the robotic arm. In this process, the movement trajectory within the time period from the starting time to the end time is recorded as the first motion trajectory, and the operator's operation standard degree is checked through the trajectory evaluation coefficient:
[0024] If the evaluation coefficient is less than or equal to 0, it is determined that the operator has hit an obstacle and can be treated as a failure;
[0025] If the evaluation coefficient is greater than zero, then it will be treated as a pass, and the student's score will be judged based on the evaluation coefficient. The higher the evaluation coefficient, the higher the score.
[0026] Preferably, obstacles are further provided in the virtual scene, and for each obstacle provided, a corresponding obstacle coordinate range is stored, wherein the obstacle includes a side wall of the frame for loading the target object.
[0027] A computer-readable storage medium includes a stored program, wherein when the program is run, the device where the computer-readable storage medium is located is controlled to implement the above-mentioned virtual scene-based robot arm motion trajectory simulation verification method.
[0028] A processor is used to run a program, wherein when the program is run, the robot arm motion trajectory simulation verification method based on a virtual scene as described above is executed.
[0029] Compared with the prior art, the present invention has the following beneficial effects:
[0030] 1. This invention provides a highly customizable training environment for robotic arm operators by creating virtual scenarios. In practice, due to equipment, space, and time constraints, operators are often exposed to only a limited number of work scenarios. This can lead to operational errors due to lack of experience when faced with complex and ever-changing real-world tasks. The creation of virtual scenarios, however, overcomes these limitations by rapidly generating a variety of complex work scenarios, including varying workspace layouts, obstacle distribution, and task objectives, tailored to specific training needs. Operators can practice repeatedly in these diverse virtual scenarios, becoming familiar with a variety of potential situations and significantly improving their adaptability. Furthermore, virtual scenarios can gradually increase the difficulty and complexity of tasks based on the operator's training progress and level of mastery, enabling personalized training plans and further enhancing training effectiveness. This diverse training approach not only helps operators better master robotic arm operation skills but also cultivates their adaptability and decision-making abilities when faced with complex tasks, laying a solid foundation for them to cope with various challenges in real-world work.
[0031] 2. The present invention compares the motion path formed by the operator operating the robotic arm in the virtual scene with the standard path to obtain the similarity, and finally obtains the evaluation coefficient by combining factors such as whether it collides with obstacles. Compared with the traditional manual judgment method, this evaluation system has significant objectivity and efficiency. In traditional training, the skill level of the operator mainly depends on the subjective judgment of the evaluator. Different evaluators may give different evaluations due to factors such as personal experience and understanding of standards, resulting in a lack of uniformity and accuracy in the evaluation results. However, this solution uses computer algorithms to accurately analyze and compare motion paths, and can provide objective and accurate evaluation results based on data, avoiding interference from human factors. At the same time, computer algorithms can complete complex calculations and analyses in a short period of time, greatly improving the efficiency of the evaluation and saving time and labor costs. In addition, the evaluation results based on the evaluation coefficient can also clearly reflect the advantages and disadvantages of the operator in terms of operational standardization, path planning rationality, and obstacle avoidance ability, providing a clear direction for subsequent targeted training. This objective and efficient evaluation system can not only accurately measure the skill level of operators, but also provide strong support for the optimization and adjustment of training plans, thereby comprehensively improving the quality of training and ensuring that operators can better master the operation skills of the robotic arm to meet the needs of actual work.
[0032] 3. This invention provides a low-cost, highly efficient solution for training robotic arm operators through the construction of virtual scenarios and evaluation using computer algorithms. Traditional training methods require significant equipment and site resources, which not only increases training costs but can also cause equipment wear and damage, shortening its service life. Virtual scenario training, on the other hand, eliminates the need for actual robotic arm equipment and complex site layouts and can be implemented solely through computer software, significantly reducing training costs. Furthermore, virtual scenarios can support simultaneous training for multiple operators, unrestricted by the number of equipment and site space, improving the efficient use of training resources. Furthermore, virtual scenario training can be flexibly scheduled based on the individual operator's progress, avoiding the situation in traditional training where some personnel have to wait or others fall behind due to a unified schedule, further improving the efficiency and effectiveness of training. By reducing training costs and improving the efficient use of training resources, this technical solution not only saves companies significant training funds but also enables more personnel to receive high-quality training, improving the operational proficiency of the entire team, thereby generating greater economic and social benefits for the company. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly describes the drawings required for the specific embodiments or the description of the prior art. Similar elements or parts are generally identified by similar reference numerals throughout the drawings. Elements or parts in the drawings are not necessarily drawn to scale.
[0034] Figure 1 is a flow chart of the method of the present invention;
[0035] Figure 2 This is a specific scenario diagram of an embodiment of the present invention. Figure 1 (starting point);
[0036] Figure 3 This is a specific scenario diagram of an embodiment of the present invention. Figure 1 (Process 1);
[0037] Figure 4 This is a specific scenario diagram of an embodiment of the present invention. Figure 1 (Process 2);
[0038] Figure 5 This is a specific scenario diagram of an embodiment of the present invention. Figure 1 (end). DETAILED DESCRIPTION
[0039] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.
[0040] It will be understood that when used in this specification and the appended claims, the terms “comprises” and “comprising” indicate the presence of described features, integers, steps, operations, elements and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups thereof.
[0041] It should also be understood that the terms used in the present specification are only for the purpose of describing particular embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, the singular forms "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise.
[0042] It should be further understood that the term "and / or" used in the present description and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
[0043] In one embodiment of the present invention, a method for simulating and verifying a robot arm's motion trajectory based on a virtual scene is provided, comprising the following steps:
[0044] Step 1: Construct a virtual scene, which at least includes a robotic arm, an obstacle, a target object, a starting point for the object's movement, a passing point for the object's movement, and an end point for the object's movement. The robotic arm is provided with a gripping end point.
[0045] For each obstacle set, a corresponding obstacle coordinate range is stored, which is used to reflect that when the time coordinates of the clamping endpoint motion trajectory fall into the obstacle coordinate range at a certain moment, it means that a collision with the obstacle occurs at that moment.
[0046] Step 2: Record the coordinates of the gripping endpoint of the robotic arm at each time point (x i ,y i ,z i ), and the time when the robot arm grabs the target object is the starting time, and the time when the robot arm places the target object at the end point of the object movement is the end time, and the target object movement trajectory (first movement trajectory) T1 = {(x1, y1, z1), (x2, y2, z2), …, (x n ,y n ,z n ),}.
[0047] During this process, it is important to ensure that the time intervals of all data points are consistent. If the time intervals are inconsistent, they can be adjusted using interpolation methods (such as linear interpolation or spline interpolation). If there is noise in the data, filtering methods (such as sliding average filtering or Gaussian filtering) can be used to smooth the trajectory.
[0048] In this step, in addition to directly operating the manipulator in the virtual scene by dragging the mouse or setting parameters and obtaining the coordinates of the manipulator's gripping endpoint, the virtual scene can also be combined with the real scene data, that is, by installing a sensor in the actual manipulator's gripping endpoint, the motion trajectory of the manipulator's gripping endpoint is obtained, and then reflected in the virtual scene, and the real transport trajectory is verified or evaluated by capturing the real motion trajectory data. That is, in this embodiment, the state of the physical manipulator and the virtual model (such as joint angle, end position) can be synchronized in real time through sensors (such as IMU, encoder) or industrial communication protocols (OPC UA, EtherCAT). It is convenient to compare the trajectory errors of the virtual and actual executions in the future, and adjust the virtual model parameters or re-plan the trajectory according to the error.
[0049] Step 3: Define a standard motion trajectory to verify whether the first motion trajectory is standard. The standard motion trajectory is an idealized motion path defined in combination with the current virtual scene and the target object, which is related to the obstacles in the virtual scene and the properties of the target object.
[0050] The standard motion trajectory is expressed as:
[0051] In this step, it is also necessary to ensure that the first motion trajectory T1 and the standard motion trajectory T s If the time points are inconsistent, they can be adjusted through interpolation. For example, if the time points of the standard trajectory are denser, the real-time trajectory is interpolated to make its time points consistent with the standard trajectory.
[0052] In addition, attention should be paid to the spatial alignment of the two motion trajectories. If the starting or ending points of the two trajectories are inconsistent, they are aligned through translation or rotation operations. That is, by calculating the center of mass of the two trajectories and translating one trajectory to the center of mass of the other trajectory.
[0053] Step 4: Calculate the similarity F(S) between the two trajectories.
[0054] In this step, the similarity is measured by calculating the Euclidean distance between the corresponding points of the two trajectories and summing them up to get the total distance. The specific operation is as follows:
[0055] S1: Calculate the Euclidean distance between corresponding points on two trajectories:
[0056]
[0057] In the formula, (x i ,y i ,z i ) represents the coordinates of time point i in the first motion trajectory, Represents the coordinates of time point i in the standard motion trajectory.
[0058] S2: Accumulate the distances of all corresponding points to get the total distance:
[0059]
[0060] Where d(i) is the Euclidean distance between two motion trajectories at time point i, and n is the total number of sampling points.
[0061] S3: Normalize the total distance D to the interval [0,1] to obtain the similarity F(S):
[0062]
[0063] Where Dmax is the maximum possible distance. In this embodiment, it is determined according to the allowable range of the trajectory, such as the workstation range.
[0064] The calculated similarity F(S) can be used to determine the degree of similarity between two trajectories. If F(S) is close to 1, the two trajectories are very similar; if F(S) is close to 0, the two trajectories are quite different.
[0065] In one embodiment of the present invention, dynamic time warping (DTW) can also be used to calculate the similarity of trajectories. DTW can handle the situation where the time points of the trajectories are inconsistent and can better capture the shape similarity of the trajectories.
[0066] Step 4: Define a trajectory evaluation coefficient, and obtain the standard degree of the first motion trajectory based on the trajectory evaluation coefficient.
[0067]
[0068] Where Z is the trajectory evaluation coefficient, F(S) is the similarity between the first motion trajectory and the standard motion trajectory, and O i is the collision coefficient of the first motion trajectory at time point i, where the collision coefficient is 1 when colliding with an obstacle and zero when not colliding with an obstacle, β i is the collision weight coefficient at time point i, α s , α o are the weight coefficients of similarity and collision coefficient respectively.
[0069] It can be seen from the above formula that the trajectory evaluation coefficient can be used to judge whether a motion trajectory is standard or high-scoring. For example, in the teaching scenario, when α s , α o , β i When both are 1, the student operates the simulated robotic arm and moves the robotic arm endpoint by clicking the mouse, etc., and then sets the movement trajectory of the robotic arm. In this process, the movement trajectory from the starting time to the end time is recorded as the first movement trajectory. Then, the trajectory evaluation coefficient can be used to see the student's operation standard. At this time, if the student encounters an obstacle, the evaluation coefficient will be less than or equal to 0. If the evaluation coefficient is not greater than zero, it means that the student will encounter an obstacle during the track design process, and it can be treated as a failure. If the evaluation coefficient is greater than zero, it means that in this process, there is a certain distance between the first movement trajectory and the obstacles in the current scene, and no obstacles in the scene are encountered. At this time, it is treated as a passing score, and the student's score is judged based on the evaluation coefficient (completely affected by similarity). The higher the evaluation coefficient, the higher the score.
[0070] In addition to the teaching scenarios described above, the evaluation coefficient can also be applied to different application scenarios. In this case, the evaluation coefficient can be adjusted by adjusting α s , α o The value of β can be used to obtain the evaluation coefficients with different degrees of influence from similarity and expansion coefficient. In addition, β can also be adjusted. i The value of β can be used to obtain the evaluation coefficients affected by the expansion coefficients at different locations or at different time points. For example, the weight β of the expansion coefficient of the obstacles around the starting point and end point or important instruments can be i Set to a larger value, the weight β at other positions in the process i Set it to a smaller value to reflect the different importance of different points.
[0071] In one embodiment of the present invention, the calculation of the evaluation coefficient also takes into account the degree of inclination of the robotic arm when picking up an object. The specific calculation formula is as follows:
[0072]
[0073] Among them, F(RS) i Indicates the tilt coefficient, which is used to reflect the standard degree of tilt at the clamping endpoint time point i, μ i Represents the tilt weight coefficient.
[0074] The tilt coefficient F(RS) is expressed by the similarity compared with the standard angle data. The acquisition process is as follows:
[0075] S01: Record the rotation angle coordinates of the gripping end point of the robotic arm at each time point (Rx i ,Ry i ,Rz i ), and take the time when the robot arm grabs the target object as the starting time, and the time when the robot arm places the target object at the end point of the object movement as the end time, and obtain the first rotation time series data of the target object's rotation in chronological order:
[0076] RT1={(Rx1,Ry1,Rz1),(Rx2,Ry2,Rz2),…,(Rx n ,Ry n ,Rz n ),}.
[0077] Similarly, in this process, it is necessary to ensure that the time intervals of all data points are consistent. If the time intervals are inconsistent, they can be adjusted through interpolation methods (such as linear interpolation or spline interpolation).
[0078] Similarly, in this step, in addition to directly operating the robotic arm in the virtual scene through various methods such as dragging the mouse or setting parameters to obtain the rotation angle coordinates of the robotic arm's gripping end point, the virtual scene can also be combined with the real scene data, that is, by installing corresponding sensors in the actual robotic arm's gripping end point, the rotation angle of the robotic arm's gripping end point can be obtained, and then reflected in the virtual scene, and the real data can be verified or evaluated by capturing the real data.
[0079] S02: Define a standard rotation scenario, i.e., standard rotation time series data, for verifying whether the first rotation time series data meets the standard. The standard rotation scenario is an idealized rotation angle defined in conjunction with the current virtual scene and the target object type. In this embodiment, a corresponding standard rotation scenario is set for each target object type. This is related to the obstacles in the virtual scene and the properties of the target object. In this embodiment, the same object may be classified as a different target object type when carrying different types or quantities of substances. For example, if the target object is a full cup of water, the standard rotation scenario is no rotation. For example, if the target object is half full or less than half full, the standard rotation scenario is a rotation scenario that allows a certain rotation angle to avoid obstacles. For another example, in cases where the starting and ending points require front-side-up reversal (starting point right side up, ending point wrong side up), the standard rotation scenario involves reversal at a predetermined optimal point (or range of points).
[0080] The standard rotation time series data is expressed as:
[0081] In this step, it is also necessary to ensure that the first rotation timing data RT1 and the standard rotation timing data RT s If the time points are inconsistent, they can be adjusted through interpolation.
[0082] S03: Calculate the similarity (tilt coefficient) F(RS) of the two rotated time series data.
[0083] In this step, the Euclidean distance between the corresponding points of the two rotated time series data is calculated and the total distance is obtained by accumulating them to measure the similarity. The specific operation is as follows:
[0084] S1: Calculate the Euclidean distance between corresponding points on two rotated time series data:
[0085]
[0086] In the formula, (x i ,y i ,zi ) represents the coordinates of time point i in the first rotation time series data, Represents the coordinates of time point i in the standard rotation time series data.
[0087] S2: Accumulate the distances of all corresponding points to get the total distance:
[0088]
[0089] Where d(i) is the Euclidean distance between two motion trajectories at time point i, and n is the total number of sampling points.
[0090] S3: Normalize the total distance D to the interval [0,1] to obtain the similarity S:
[0091]
[0092] Where, RD max is the maximum possible distance. In this embodiment, it is determined based on the allowable tilt of the item type, for example, the rotation angle Rx about the X axis must not exceed 180°.
[0093] The calculated similarity F(RS) can be used to determine the degree of similarity between two trajectories. If F(RS) is close to 1, the two trajectories are very similar; if F(RS) is close to 0, the two trajectories are quite different.
[0094] In one embodiment of the present invention, the following steps are further included:
[0095] Step 5: Obtain a trajectory evaluation coefficient, set at least one alarm threshold range, and perform corresponding alarm operations on the trajectory evaluation coefficient evaluated in real time based on the range.
[0096] In one embodiment of the present invention, in addition to extracting the coordinates of the gripping endpoints of the manipulator and obtaining the evaluation coefficients based on similarity and obstacle collision, the above evaluation coefficients also obtain other points of the manipulator (such as a bearing turning point, etc.), which are specifically expressed as follows:
[0097]
[0098] Where Z j is the evaluation coefficient of the j-th robotic arm point, with a total of m points. j Any calculation method in the above embodiments may be used.
[0099] In one embodiment of the present invention, the following steps are further included:
[0100] Step 6: Compare the trajectory errors between the virtual and actual executions, and adjust the virtual model parameters or replan the trajectory based on the errors.
[0101] like Figure 2-Figure 5 As shown, the following is an example of scene construction, the following is the key code:
[0102] 1. Coordinate codes of each joint of the robotic arm
[0103]
[0104]
[0105]
[0106]
[0107]
[0108]
[0109] 2. Grab and place item code
[0110]
[0111]
[0112]
[0113] In summary, this invention, through virtual scenarios and computer algorithm evaluation, provides an efficient, objective, and low-cost solution for robotic arm operator training, effectively addressing the shortcomings of traditional training. Traditional training, limited by equipment and venues, struggles to provide a diverse and personalized training environment, leading to operators' lack of experience and prone to errors when faced with complex tasks. Virtual scenarios, on the other hand, can quickly generate complex work scenarios with varying layouts, obstacle distributions, and task objectives, allowing operators to practice repeatedly and improve their adaptability and resilience. Furthermore, virtual scenarios can gradually increase task difficulty based on training progress and mastery, enabling personalized training plans and enhancing training effectiveness. Traditional training relies on subjective judgment by evaluators, resulting in inconsistent and inaccurate evaluation results due to varying personal experience and understanding of standards. This solution compares the operator's robotic arm motion path in a virtual scenario with the standard path, deriving a similarity and then factoring in factors such as collisions to generate an evaluation coefficient. This computer algorithm-based evaluation system is data-based, objective, and accurate, avoiding human interference. It can also rapidly complete complex calculations and analysis, improving evaluation efficiency and saving time and labor costs. The evaluation results can also clearly reflect the strengths and weaknesses of operators in terms of operational standardization, path planning rationality and obstacle avoidance capabilities, providing direction for targeted training.
[0114] Those skilled in the art will appreciate that the units of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.
[0115] In the embodiments provided by the present invention, it should be understood that the division of units is merely a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.
[0116] In addition, the functional units in the various embodiments of the present invention may be integrated into a single processing unit, each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.
[0117] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk, and other media that can store program codes.
[0118] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and description of the present invention.
Claims
1. A method for simulating and verifying the motion trajectory of a robotic arm based on a virtual scene, characterized in that: The following steps are involved: Constructing a virtual scene, wherein the virtual scene at least includes a robotic arm, a target object, a starting point for the object's movement, and an end point for the object's movement, wherein the robotic arm is provided with a gripping end point; Record the coordinates of the gripping endpoint of the robotic arm at each time point (x i ,y i ,z i ), according to the time sequence, we get the first motion trajectory T1 = {(x1, y1, z1), (x2, y2, z2), …, (x n ,y n ,z n ),}; defining a standard motion trajectory and calculating its similarity with the first motion trajectory; Calculate the trajectory evaluation coefficient as follows: Where F(S) is the similarity between the first motion trajectory and the standard motion trajectory, O i is the collision coefficient of the first motion trajectory at time point i, β i is the collision weight coefficient at time point i, α s , α o are the weight coefficients of similarity and collision coefficient respectively.
2. The method for simulating and verifying the motion trajectory of a robotic arm based on a virtual scene according to claim 1, wherein: The calculation of the evaluation coefficient also takes into account the degree of inclination of the robot arm when picking up an object. The specific calculation formula is as follows: Among them, F(RS) i Indicates the tilt coefficient, which is used to reflect the standard degree of tilt at the clamping endpoint time point i, μ i represents the tilt weight coefficient at time point i, α RS Indicates the weight coefficient of the tilt coefficient.
3. The method for simulating and verifying the motion trajectory of a robotic arm based on a virtual scene according to claim 2, wherein: The tilt coefficient F(RS) is expressed by the similarity compared with the standard angle data.
4. The method for simulating and verifying the motion trajectory of a robotic arm based on a virtual scene according to claim 3, wherein: The process of obtaining the tilt coefficient F(RS) is as follows: Record the rotation angle coordinates of the gripping end point of the robot arm at each time point (Rx i ,Ry i ,Rz i ), and obtaining first rotation time series data of the rotation of the target object in chronological order; defining standard rotation time series data for verifying whether the first rotation time series data meets the standard; The similarity between the first rotation time series data and the standard rotation time series data is calculated as the tilt coefficient F(RS).
5. The method for simulating and verifying the motion trajectory of a robotic arm based on a virtual scene according to claim 4, characterized in that: For each target item type, at least one corresponding standard rotation situation is set.
6. The method for simulating and verifying the motion trajectory of a robotic arm based on a virtual scene according to claim 1, characterized in that: Weight coefficient α of similarity and collision coefficient s , α o The value of is 1.
7. The method for simulating and verifying the motion trajectory of a robotic arm based on a virtual scene according to claim 6, characterized in that: The operator operates the simulated robotic arm and moves the endpoints of the robotic arm by clicking the mouse, thereby setting the movement trajectory of the robotic arm. In this process, the movement trajectory from the starting time to the end time is recorded as the first motion trajectory, and the operator's operation standard is checked through the trajectory evaluation coefficient: If the evaluation coefficient is less than or equal to 0, it is determined that the operator has hit an obstacle and can be treated as a failure; If the evaluation coefficient is greater than zero, then it will be treated as a pass, and the student's score will be judged based on the evaluation coefficient. The higher the evaluation coefficient, the higher the score.
8. The method for simulating and verifying the motion trajectory of a robotic arm based on a virtual scene according to claim 1, wherein: Obstacles are also set in the virtual scene, and for each obstacle set, a corresponding obstacle coordinate range is stored. The obstacle includes the side wall of the frame where the target object is loaded.
9. A computer-readable storage medium, characterized in that The computer-readable storage medium includes a stored program, wherein when the program is running, the device where the computer-readable storage medium is located is controlled to execute the virtual scene-based robotic arm motion trajectory simulation verification method according to any one of claims 1 to 8.
10. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes the robot arm motion trajectory simulation verification method based on a virtual scene according to any one of claims 1 to 8.