A cloud-edge collaborative mobile accessibility experience and evaluation method

Abstract

A cloud-edge collaborative method for mobile accessibility experience and evaluation includes the following steps: First, the requester initiates a task, and the edge automatically traverses designated mobile applications that have enabled accessibility assistive technologies to obtain accessibility structure information. Then, the cloud, based on the relevant requirements of accessibility standards, generates corresponding experience and evaluation crowdsourcing tasks and their corresponding functional links. Next, disabled or able-bodied users apply for the task on a browser client. After the task is bound, disabled or able-bodied users remotely control a robotic arm on the interface to access the mobile application on the testing machine and receive real-time voice and video streaming feedback from the device to achieve compliance evaluation and submit the results. Finally, all results are integrated to output a report.

Description

Technical fields:

[0001] This invention relates to a cloud-edge collaborative mobile accessibility experience and evaluation method, belonging to the field of computer technology and software. Background technology:

[0002] The term "accessibility" originates from the English word "accessibility," initially referring to providing accessible products, equipment, services, and environments for people with disabilities. In the 2000 Okinawa Charter for a Global Information Society, the concept of "information accessibility" was formally introduced, meaning that anyone (whether able-bodied or disabled, young or old) can understand, interact with, and utilize information equally and conveniently in any situation. With the rapid development of mobile internet, most online services have now fully shifted to mobile applications, and research on information accessibility has also focused on mobile application accessibility. Statistics show that the total number of smartphone users worldwide reached 3.9 billion in 2021, a year-on-year increase of 6.1%. It can be said that mobile internet has become an indispensable part of people's daily lives. Meanwhile, according to estimates from the World Health Organization, approximately 285 million people worldwide have some degree of visual impairment, of whom about 39 million are completely blind. In addition, approximately 466 million people suffer from hearing loss, representing approximately 3.7% and 6.1% of the global population, respectively. Clearly, how to help people with disabilities integrate into the information society, bridge the "information gap" of the mobile internet, and enable all people to enjoy the convenience of the mobile internet equally and conveniently is one of the important issues in building an inclusive society.

[0003] The assessment of accessibility compliance for mobile applications by disabled or able-bodied users is an important method to help optimize the accessibility support of applications. The main challenges are as follows: (1) How to achieve automated mobile application traversal to obtain complete application accessibility structure information for generating experience and assessment tasks. (2) How to combine accessibility standard requirements with the assessment process so that the assessment results are more instructive for accessibility developers. (3) How to achieve support for disabled or able-bodied users to remotely control mobile devices and receive device feedback in real time from both hardware and software perspectives. Summary of the Invention:

[0004] To address the above problems and challenges, this invention proposes a cloud-edge collaborative mobile accessibility experience and evaluation method.

[0005] The specific functional steps of a cloud-edge collaborative mobile accessibility experience and evaluation method are as follows: S101: The requester creates an experience and evaluation task for a specified mobile application and specifies the scope of accessibility standard items to be evaluated. S102: The requester starts the created task, and the cloud server sends the task information to the edge device. The edge device automatically downloads and installs the specified application on the mobile device according to the user configuration. S103: The edge device controls the test machine to enable accessibility mode (Talkback on Android, Voiceover on iOS), and the host computer controls the test machine to automatically traverse the specified mobile application and obtain application structure information. S104: The cloud server generates experience and evaluation tasks in batches based on the application structure data obtained from the edge device traversal and the user configuration, along with the specified accessibility standard items and their related application function links. At this point, the pre-generation of experience and evaluation tasks is complete. S105: Disabled or able-bodied users open the specified page from a computer browser client and apply for experience and evaluation tasks according to the interface process. S106: The cloud server receives the task request from the tester, randomly assigns the experience and evaluation task, and binds it to the edge device. S107: The edge device receives the task information, operates the robotic arm to access the mobile application, opens the corresponding page according to the specified link, and enters a ready state, awaiting remote operation from the tester. S108: Disabled or able-bodied users receive the edge device's ready signal, obtain the standard items and methods to be evaluated through browser voice reading, remotely operate the robotic arm via keyboard to experience and access the specified application on the test device, obtain feedback information from the test device application through browser voice reading, fill in and submit the evaluation results and feelings. S109: After receiving the results of all pre-generated experience and evaluation tasks, the cloud server integrates the data and generates a report. S110: After completing the evaluation task, the user requesting the experience and evaluation task can view and download the evaluation report on the web page.

[0006] Specifically, the information required when creating the evaluation task for the specified mobile application in step S101 includes: S201: mobile application name, applicant organization name, mobile application type (Android or iOS), mobile application download method, mobile application download link, test rule scope, test login account, test login password, and test instructions. The test rule scope needs to be determined by selecting several items from accessibility standards such as "Technical Requirements and Test Methods for Internet Content Accessibility" (GB / T 37668—2019) or Web Content Accessibility Guidelines (WCAG) 2.1. The test login account, test login password, and test instructions are optional.

[0007] Specifically, in step S102, the edge device consists of the following devices and functions: S301: The edge device hardware includes a host computer chassis, a robotic arm, and a testing machine; S302: The host computer functions to receive task information, download the specified mobile application installation package, install the mobile application to the testing machine, operate the robotic arm to access the mobile application, capture the mobile application page and save the screenshot to the cloud server, automatically traverse the mobile application, and support pushing the screen recording and audio information of the mobile application to the cloud server; S303: The robotic arm functions to simulate the left swipe, right swipe, up swipe, down swipe, single click at specified coordinates, and multi-click at specified coordinates when a disabled user uses the mobile application, thereby enabling access to the mobile application; S304: The testing machine functions to receive control signals from the host computer, realize access control in barrier-free mode, and simultaneously record the screen and push the audio and video to the cloud server in real time.

[0008] Specifically, step S103, the automatic traversal process of the specified mobile application on the edge includes single-page sequential traversal and multi-page breadth-first traversal, while also supporting the handling of waterfall layout or focus traps: S401: Traversing a single page requires not only traversing all controls on a certain page, but also identifying or obtaining the corresponding attribute information of the controls. Specifically, as follows: Upon entering the page, the focus box is located at the first focusable control on the current page. The audiovisual intelligent algorithm and other modules are called to obtain and record the image or semantic attributes such as the coordinates of the focus box. Then, the robotic arm performs a "right swipe" operation on the screen, at which point the focus box on the page will slide to the next focusable control, and the corresponding algorithm is called again to obtain the attribute information of the new focus box. Ideally, by repeating the above steps to move the focus box from the first control to the last control, the sequential traversal of the page can be completed; S402: Based on single-page traversal, the breadth-first traversal of the full mobile application for multiple pages mainly includes the following basic steps: (1) Enter a page of the mobile application, initially the home page, and perform single-page traversal: continuously swipe right, and during this period, judge and handle the waterfall flow or focus trap until the set swipe number threshold is reached, the traversal ends, and all focus boxes of the home page are added to the breadth-first traversal queue. (2) When the breadth-first traversal queue is not empty, take the first focus box from it and remove it from the queue. Since the removed focus box may not be on the current page, first calculate the path from the current page to the target focus box, and reach the corresponding focus box according to this path. After clicking the focus box, judge whether a brand new page has been entered. If so, repeat (1); otherwise, remove the next focus box from the queue and repeat (2) until the queue is empty. (3) The page layer number at which the traversal stops can be preset. If the current page layer number is less than the preset traversal stop layer number, the traversal process continues; otherwise, the traversal stops. S403: During the above sequential traversal process, there may be pages with a waterfall layout. Waterfall layouts dynamically load new elements when the user focuses on the last control, resulting in nearly infinitely many focusable controls being continuously loaded. This is common in short video mobile applications and online e-commerce mobile applications. Such cases require special handling; otherwise, the traversal cannot be stopped. The specific processing flow is as follows: (1) During the traversal, each time the robotic arm slides to the right, a new focus box object is generated, and the corresponding attributes are identified or obtained. After numbering the focus box using an auto-incrementing method, the newly generated focus box object is added to the focus box list of this page. (2) If the number of right slides reaches the preset threshold, it is necessary to determine whether the page is a waterfall layout page. Since the accessibility operation modes of each system are slightly different, special adaptation is required here. For iOS, if the focus box's position and content remain unchanged and it's located in the bottom right corner of the screen (i.e., the focus box is fixed as the last control on the page and doesn't change), then the page traversal is complete, and it's a normal page. For Android, if the page returns to the first focus box after a right swipe and starts looping again, then it's a normal page. If none of the above conditions are met, then it's considered a waterfall layout page.(3) If the page is not a waterfall layout, it will end normally; if it is a waterfall layout, it will delete the duplicate controls and then end; S404: During the mobile application traversal, the focus may be stuck due to poor accessibility support of the application. There are two types of focus stuck: (1) single-point stuck, the focus box stays on the same control and cannot be moved to other control positions no matter what operation is performed; (2) sequence stuck, the focus box stays in a limited sequence of controls and cannot be moved to other control positions outside the sequence no matter what operation is performed. The solution is as follows: (1) Single-point stuck judgment: If the difference between the upper left corner coordinates, width and length of the two focus boxes is within the set allowable error range, the two focus boxes are judged to be the same, otherwise they are judged to be different. Set a number of consecutive identical focus boxes. When the number of consecutive identical focus boxes reaches this threshold, it is judged to be a focus stuck. (2) Sequence stuck judgment: When the number of times a focus box appears in a fixed interval (greater than 1) exceeds the set maximum number of repetitions, it is judged to be a sequence focus stuck. (3) Traps: If it is a single-focus trap, delete the last consecutive identical focus boxes in the focus box list and keep only one. If it is a sequence-focus trap, delete the last consecutive identical cycle in the focus box list and keep only one cycle.

[0009] Specifically, step S103 applies the following structural information: S501: applying screenshots of each page and their navigation relationships, screenshots of each control on each page, accessible assistance mode text reading, control focus box position coordinates, and control order relationships;

[0010] Specifically, step S104, the process of batch pre-generating the experience and evaluation tasks of the specified accessibility standard items and their related application function links, is as follows: S601: Select a standard item that the user configures to be evaluated, randomly extract a page from the application structure obtained in step S103, form a mapping relationship with the item to construct an experience and evaluation task, and repeat the above process to generate the evaluation tasks corresponding to all user-configured evaluation items.

[0011] Specifically, step S105, the task application process for disabled or able-bodied users includes: S701: (1) applying for a testing task; (2) waiting for the connection of the robotic arm and the mobile phone; (3) automatically locating the error page; (4) obtaining control of the robotic arm;

[0012] Specifically, step S106, the steps of binding the cloud server to the experience and evaluation task to the edge include: S801: (1) Sending the task information to the edge and starting the task process; (2) Locking the edge and no longer receiving new tasks; (3) Automatically releasing the edge after the task is completed.

[0013] Specifically, step S107, opening the corresponding page according to the specified link, includes: S901: (1) Obtain the unique identifier of the starting page and the unique identifier of the target page; (2) Calculate the page jump path instruction queue; (3) Dequeue one step from the page jump path instruction queue to obtain the jump instruction; (3) If the instruction is to go back one page, try to go back using the instruction of accessibility assistance technology. (3.1) If the back instruction runs successfully, determine whether to go back to the correct page; (3.1.1) If the back is successful, update the current page information, continue to dequeue instructions until the queue is empty, the program ends normally, and the corresponding page is successfully opened according to the specified link; (3.1.2) If the back fails, restart the application, and the starting page is the home page, repeat (2) and the subsequent steps. (3.2) If the back instruction runs unsuccessfully, try to detect the back button through a visual algorithm. If the button exists, click back and go to (3.1); if the button does not exist, go to (3.1.2). (4) If the instruction is not a back, the target page and target focus box information are read from the jump instruction. (4.1) If the current page is the target page, the focus box coordinates are used to directly double-click the focus box to jump to the next page, go to (3.1.1); (4.2) If the current page is not the target page, the target focus box is not detected, the program is terminated directly, and the corresponding page is opened according to the specified link but fails.

[0014] Specifically, in step S108, the method for disabled or able-bodied users to remotely operate the robotic arm to access the application via the keyboard is as follows: S101: Remotely control the robotic arm via the keyboard keys, press the w key to make the robotic arm control the phone to return to the page, press the s key to make the robotic arm control the phone to double-click to enter the page, press the a key to make the robotic arm control the phone to swipe left, press the d key to make the robotic arm control the phone to swipe right, and press the k key to turn on the sound from the phone.

[0015] Specifically, in step S108, the evaluation results and feelings filled in and submitted by disabled or able-bodied users include: S1101: whether the evaluated accessibility item is passed, and if not, a brief description of the reason for the error;

[0016] Specifically, in step S109, the report generated by the cloud server includes: S1201: the level of each standard item, pass status, error examples, error jump paths, and a page-by-page errata table of all errors, etc.

[0017] Specifically, in step S110, the report format that can be downloaded only if the experience and evaluation task requires includes: S1301: PDF format document or Microsoft Office Word format document.

[0018] In summary, this invention creates a cloud-edge collaborative mobile accessibility experience and evaluation method, which has the following beneficial effects: (1) It realizes automated mobile application accessibility mode traversal, thereby obtaining complete application structure information; (2) Then, it combines the structure information with accessibility standard requirements to generate experience and evaluation tasks with accessibility items, thereby enhancing the guiding value of the evaluation results for accessibility developers; (3) It realizes support for disabled or able-bodied users to remotely control mobile devices and receive device feedback in real time from the hardware and software level, truly opening up the functional link for remote experience and evaluation of disabled or able-bodied users. Attached image description:

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 is a schematic diagram of the workflow of a cloud-edge collaborative mobile accessibility experience and evaluation method provided by the present invention;

[0021] Figure 2 is a device prototype diagram of a cloud-edge collaborative mobile accessibility experience and evaluation method provided by the present invention;

[0022] Figure 4 is a schematic diagram of the actual edge hardware device provided by the present invention;

[0023] Figure 3 is a schematic diagram of the black focus box during the traversal of the accessibility mode traversal algorithm provided by the present invention (the focus box is green for Android).

[0024] Figure 5 is a schematic diagram of the functional flow of the edge-end robotic arm provided by the present invention, which simulates a disabled user controlling a mobile device by sliding and clicking.

[0025] Figure 6 is a schematic diagram of the function provided by the present invention, which allows disabled or able-bodied users to remotely control a robotic arm by specifying keys on a keyboard, receive real-time information feedback from a mobile device via audio, and submit their experience on the platform.

[0026] Figure 7 is a schematic diagram of the algorithm flow for opening the corresponding page by a specified link provided by the present invention. Specific implementation methods:

[0027] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0028] This example uses the accessibility and compliance experience and evaluation task of a certain APP as an example, and includes the following specific steps:

[0029] S101: The requester creates an experience and evaluation task for a specified mobile application, specifying the scope of accessibility compliance testing standards;

[0030] S102: The requester initiates the task, and the cloud server sends the task information to the edge device. The edge device automatically downloads and installs the specified application on the mobile device according to the configuration.

[0031] S103: The edge control test device is enabled in accessibility mode (Talkback for Android, Talkback for iOS), and the host computer controls the test device to automatically traverse the specified mobile application and obtain the application structure information.

[0032] S104: Based on the application structure data and user configuration obtained from edge computing, the cloud server generates batch experience and evaluation tasks for specified accessibility standard items and their related application function links. At this point, the pre-generation of experience and evaluation tasks is complete.

[0033] S105: Disabled or able-bodied users can open the designated page from a computer browser client and apply for the experience and evaluation task according to the interface process.

[0034] S106: The cloud server receives a request from a tester to apply for a task, randomly assigns test and evaluation tasks, and binds them to the edge.

[0035] S107: The edge receives the task information, operates the robotic arm to access the mobile application, opens the corresponding page according to the specified link, and enters the ready state, waiting for the tester to operate remotely.

[0036] S108: When a disabled or able-bodied user receives the edge ready signal, he / she obtains the standard items and methods to be evaluated through the browser's voice reading, then remotely operates the robotic arm through the keyboard to experience and access the specified application on the test machine, and obtains the application feedback information of the test machine through the browser's voice reading, and fills in and submits the evaluation results and feelings.

[0037] S109: After receiving the results of all pre-generated experience and evaluation tasks, the cloud server integrates the data and generates a report;

[0038] S110: After the user completes the evaluation task, the user can view and download the evaluation report on the webpage.

Claims

1. A cloud-edge collaborative mobile accessibility experience and evaluation method, characterized in that Includes the following steps: S101: The requester creates an experience and evaluation task for a specified mobile application and specifies the scope of accessibility standard items to be evaluated. S102: The requester initiates the task, and the cloud server sends the task information to the edge device. The edge device automatically downloads and installs the specified application on the mobile device according to the user configuration. S103: The edge control test machine starts the accessibility mode, and the host computer controls the test machine to automatically traverse the specified mobile application and obtain the application's accessibility structure information. S104: The cloud server generates experience and evaluation tasks for specified accessibility standard items and related application function links in batches based on the application structure data and user configuration obtained from edge traversal; at this point, the pre-generation of experience and evaluation tasks is completed. S105: Disabled or able-bodied users can open the designated page from a computer browser client and apply for the experience and evaluation task according to the interface process. S106: The cloud server receives a request from an experience user to apply for a task, randomly assigns a task, and binds it to the edge. S107: The edge receives the task information, operates the robotic arm to access the mobile application, opens the corresponding page according to the specified link, and enters the ready state, waiting for the tester to operate remotely. S108: When a disabled or able-bodied user receives the edge ready signal, he obtains the standard items and methods to be evaluated through the browser's voice reading, then remotely operates the robotic arm through the keyboard to experience and access the specified application on the test machine, and obtains the application feedback information of the test machine through the browser's voice reading, and fills in and submits the evaluation results and experience. S109: After receiving the results of all pre-generated experience and evaluation tasks, the cloud server integrates the data and generates a report; S110: After the user completes the evaluation task, the user can view and download the evaluation report on the webpage. 2.The cloud-edge collaborative mobile accessibility experience and evaluation method of claim 1, wherein: The information required when creating the evaluation task for the specified mobile application in step S101 includes: S201: Mobile application name, applicant organization name, mobile application type, mobile application download method, mobile application download link, test rule scope, test login account, test login password, and test instructions; the test rule scope needs to be determined by selecting several accessibility standards, while the test login account, test login password, and test instructions are optional.

3. The cloud-edge collaborative mobile accessibility experience and evaluation method of claim 1, wherein: In step S102, the edge consists of the following devices and their functions: S301: The edge hardware device consists of a host computer chassis, a robotic arm, and a testing machine; S302: The host computer functions to receive task information, download the specified mobile application installation package, install the mobile application to the test machine, operate the robotic arm to access the mobile application, capture the mobile application page and save the screenshot to the cloud server, automatically traverse the mobile application, and support pushing the screen recording and audio information of the mobile application to the cloud server. S303: The robotic arm functions include mimicking the left swipe, right swipe, up swipe, down swipe, single click at specified coordinates, and multi-click at specified coordinates operations of disabled users when using mobile applications, thereby enabling access to mobile applications; S304: The test unit's functions include receiving control signals from the host computer, enabling access control in barrier-free mode, recording the screen, and streaming the audio and video in real time to the cloud server and then pushing it to the tester's client.

4. The cloud-edge collaborative mobile accessibility experience and evaluation method of claim 1, wherein: Step S103, the automatic traversal process of the specified mobile application on the edge includes single-page sequential traversal and multi-page breadth-first traversal, and also supports handling waterfall layout or focus trapping. The specific method is as follows: S401: Traversing a single page requires not only traversing all controls on the page but also identifying or obtaining the corresponding attribute information of the controls. Specifically: Upon entering the page, the accessibility mode focus box is located at the first focusable control on the current page; the audiovisual intelligent algorithm module is called to obtain and record the image or semantic attributes of the focus box; then, the robotic arm performs a "right swipe" operation on the screen, at which point the focus box on the page will slide to the next focusable control, and the corresponding algorithm is called again to obtain the attribute information of the new focus box; ideally, repeating the above process to move the focus box from the first control to the last control completes the sequential traversal of the page. S402: Based on single-page traversal, the multi-page breadth-first traversal of a complete mobile application mainly includes the following basic steps: (1) Enter a certain page of the mobile application, initially the home page, and perform single-page traversal: continuously swipe right, and judge and handle the waterfall flow or focus trap during the process until the set swipe number threshold is reached, the traversal ends, and all focus boxes of the home page are added to the breadth-first traversal queue; (2) When the breadth-first traversal queue is not empty, take the first focus box from it and remove it from the queue. Since the removed focus box may not be on the current page, first calculate the path from the current page to the target focus box, and reach the corresponding focus box according to this path; After clicking the focus box, determine whether a brand new page has been entered. If so, repeat (1); otherwise, remove the next focus box from the queue and repeat (2) until the queue is empty; (3) The number of page layers to stop traversing can be preset. If the current page layer is less than the set number of page layers to stop traversing, the traversal process continues; otherwise, the traversal stops. S403: During the above sequential traversal, there may be pages with a waterfall layout; the waterfall layout will dynamically load new elements when the user focuses on the last control, resulting in nearly infinitely many focusable controls being continuously loaded, which is common in short video mobile applications and online e-commerce mobile applications; such cases need to be specially handled, otherwise the traversal cannot be stopped; the specific processing flow is as follows: (1) During the traversal, the robotic arm generates a new focus box object every time it slides to the right, identifies or obtains the corresponding attributes; after numbering the focus box in an auto-incrementing manner, the newly generated focus box object is added to the focus box list of this page; (2) If the number of right swipes reaches the preset threshold, it is necessary to determine whether the page is a waterfall page. Since the accessibility operation mode of each system is slightly different, special adaptation is required here. For iOS, if the position of the focus box and the content in the box remain unchanged and it is located in the last control in the lower right corner of the screen, the traversal is completed and it is a normal page. For Android, if it returns to the first focus box after a right swipe and starts to loop through the page, it is a normal page. If the above conditions are not met, it is determined to be a waterfall page. (3) If the page is not a waterfall page, it ends normally. If it is a waterfall page, the duplicate controls are deleted before it ends. S404: During the process of mobile application traversal, the focus trapping problem may occur due to poor accessibility support of the application. The focus trapping is divided into two cases: (1) single-point trapping, the focus box stays on the same control, and no matter how it is operated, it is impossible to move the focus box to other control positions; (2) sequence trapping, the focus box stays in a limited sequence of controls, and no matter how it is operated, it is impossible to move the focus box to other control positions outside the sequence. The handling solutions are as follows: (1) Single-point trapping judgment: If the difference between the upper left corner coordinates, width and length of two focus boxes is within the set allowable error range, the two focus boxes are judged to be the same, otherwise they are judged to be different; a number of consecutive identical focus boxes is set, and when the number of consecutive identical focus boxes reaches this threshold, it is judged to be a focus trapping; (2) Sequence trapping judgment: When the number of times a focus box appears in a fixed interval exceeds the set maximum number of repetitions, it is judged to be a sequence focus trapping; (3) Trapping handling: If it is a single focus trapping, delete the last consecutive identical focus box in the focus box list and keep only one; if it is a sequence focus trapping, delete the last consecutive identical cycle in the focus box list and keep only one cycle.

5. The cloud-edge collaborative mobile accessibility experience and evaluation method of claim 1, wherein: Step S103, applying structural information includes: S501: Screenshots of each page and their navigation relationships, screenshots of each control on each page, text read aloud in accessibility mode, coordinates of the control focus box position, and the order of controls.

6. The cloud-edge collaborative mobile accessibility experience and evaluation method according to claim 1, characterized in that: The method for batch pre-generating the experience and evaluation tasks of specified accessibility standard items and their related application function links in step S104 is as follows: S601: Select a standard item that needs to be evaluated according to user configuration, randomly extract a page from the application structure obtained in step S103, form a mapping relationship with the item to construct an experience and evaluation task, and repeat the above process to generate evaluation tasks corresponding to all user-configured evaluation items.

7. The cloud-edge collaborative mobile accessibility experience and evaluation method according to claim 1, characterized in that: Step S105, the task application process for disabled or able-bodied users, includes: S701: (1) Apply for a testing task; (2) Wait for the robotic arm and mobile phone to connect; (3) Automatically locate the error page; (4) Obtain control of the robotic arm.

8. The cloud-edge collaborative mobile accessibility experience and evaluation method according to claim 1, characterized in that: Step S106, the step of binding the cloud server experience and evaluation task to the edge, includes: S801: (1) Send the task information to the edge and start the task process; (2) Lock the edge and stop receiving new tasks; (3) Release the edge automatically after the task is completed.

9. The cloud-edge collaborative mobile accessibility experience and evaluation method according to claim 1, characterized in that: Step S107, opening the corresponding page according to the specified link, includes: S901: (1) Obtain the unique identifier of the starting page and the unique identifier of the target page; (2) Calculate the page jump path instruction queue; (3) Dequeue one step from the page jump path instruction queue to obtain the jump instruction; (3) If the instruction is to go back one page, try to go back using the instruction of accessibility assistance technology; (3.1) If the back instruction runs successfully, determine whether it has gone back to the correct page; (3.1.1) If the back is successful, update the current page information, continue to dequeue instructions until the queue is empty, the program ends normally, and the corresponding page is opened successfully according to the specified link; (3.1.2) If the back fails, restart the application and return to the starting page. For the home page, repeat (2) and the following steps; (3.2) If the back instruction fails, try to detect the back button through a visual algorithm. If the button exists, click back and go to (3.1); if the button does not exist, go to (3.1.2); (4) If the instruction is not back, read the target page and target focus box information from the jump instruction; (4.1) If the current page is the target page, double-click the focus box directly to jump to the next page through the focus box coordinates and go to (3.1.1); (4.2) If the current page is not the target page, it is determined that the target focus box is missed and the program is terminated directly. Opening the corresponding page according to the specified link fails.

10. The cloud-edge collaborative mobile accessibility experience and evaluation method according to claim 1, characterized in that: In step S108, the method for disabled or able-bodied users to remotely operate the robotic arm via a keyboard to access the application is as follows: S101: Remotely control the robotic arm via the keyboard keys. Press the W key to make the robotic arm control the phone to return to the page, press the S key to make the robotic arm control the phone to double-tap to enter the page, press the A key to make the robotic arm control the phone to swipe left, press the D key to make the robotic arm control the phone to swipe right, and press the K key to turn on the sound from the phone.

11. The cloud-edge collaborative mobile accessibility experience and evaluation method according to claim 1, characterized in that: In step S108, the evaluation results and feelings filled out and submitted by disabled or able-bodied users include: S1101: Whether the assessed accessibility item passed or failed. If not, briefly describe the reason for the error.

12. The cloud-edge collaborative mobile accessibility experience and evaluation method according to claim 1, characterized in that: In step S109, the report generated by the cloud server includes the following: S1201: Level, content and pass / fail status of each standard item, error examples, error jump paths and page-by-page errata for all errors.

13. The cloud-edge collaborative mobile accessibility experience and evaluation method according to claim 1, characterized in that: In step S110, the report format that can be downloaded only for experience and evaluation tasks includes: S1301: PDF or Word document.

Images