Method and apparatus for determining a lock type

Abstract

A method for determining the lock type of a lock. In this method, a set of original data (220, 222) of a set of locks (120, 122) is collected from a set of sensors (210, 212) respectively. The set of locks (120, 122) is installed on an object (110) at a set of positions respectively, and the set of locks (120, 122) belongs to at least one lock type among multiple lock types. A set of probability distributions (230, 232) is obtained based on the set of original data (220, 222). The probability distributions in the set of probability distributions (230, 232) are associated with the probability of the locks in the set of locks (120, 122) and the lock types to which the locks belong. The lock type (240) of the lock is determined based on the set of probability distributions (230, 232). Using this method, the performance and accuracy of lock type determination can be greatly improved.

Description

Technical Field

[0001] Example embodiments of the present disclosure generally relate to type determination, and more particularly to a method, apparatus, computer system, computer-readable medium, and system for determining a lock type of a lock for fixing a container object. Background Art

[0002] Nowadays, shipping services play an important role in people's daily lives. Countless containers are loaded onto ships every day and transported around the world. To ensure that the containers can remain relatively stationary during transportation, locks (such as twist locks) are used to fix multiple containers together. Since the locks are usually owned by shipping companies, when the containers reach their destinations, the containers need to be unloaded from the ship and all the locks need to be removed from the containers.

[0003] Several solutions for removing locks by a robotic system have been proposed. However, different types of locks are removed in different ways, so the lock type of the lock needs to be determined before removal. Currently, human workers are hired to determine the lock type, so the costs in terms of time and manpower are very high. In addition, even experienced workers may make mistakes due to the deformation and corrosion of the locks. Therefore, it is desirable to be able to determine the lock type of the lock in a more efficient and convenient manner. Summary of the Invention

[0004] Example embodiments of the present disclosure provide a solution for determining the lock type of a lock.

[0005] In a first aspect, an example embodiment of the present disclosure provides a method for determining a lock type of a lock. The method includes: collecting a set of raw data of a set of locks from a set of sensors, the set of locks being respectively mounted to an object at a set of positions, the set of locks belonging to at least one lock type among a plurality of lock types; obtaining a set of probability distributions based on the set of raw data, the probability distributions in the set of probability distributions being associated with the probability of the locks in the set of locks and the lock types to which the locks belong; and determining the lock type of the lock based on the set of probability distributions. Through these embodiments, the lock type of the lock can be determined automatically based on the raw data collected from a set of sensors without manual intervention. Therefore, compared with manual operation, the performance and accuracy can be greatly improved. In addition, since container objects are usually fixed using the same type of locks, determining the lock type based on a set of raw data of a set of locks can eliminate potential errors caused by lock deformation and erosion.

[0006] In some embodiments of the present disclosure, the locking device includes a twist lock and the object includes a container. Additionally, the robotic system can remove the locking device from the object based on the lock type indication. Generally, a large ship carries thousands of containers at a time, so countless twist locks are required to fix these containers together. Through these embodiments, the lock type can be automatically identified, and then the locking device can be removed by the robotic system during the unloading process without any manual intervention. Additionally, the number of manual workers can be significantly reduced and the efficiency of the unloading process can be greatly improved.

[0007] In some embodiments of the present disclosure, the sensors in the set of sensors include image measurement cameras, and the raw data includes the image data of the locking device. In these embodiments, sensors such as 2D cameras can be adapted to collect the image data of the locking device. Additionally, the collected image data can be used for further processing to determine the lock type to which the locking device belongs. Nowadays, 2D cameras are inexpensive and widely used for monitoring the unloading process, so these embodiments provide an effective and efficient solution for reusing sensors.

[0008] In some embodiments of the present disclosure, the sensors in the set of sensors include dimensional measurement cameras, and the raw data includes the dimensional data of the locking device. Through these embodiments, sensors such as 3D cameras can be adapted to collect the dimensions of the locking device. Specifically, the laser device in the 3D camera can measure the distance between the laser device and almost every point on the surface of the locking device. Therefore, the point cloud data of the locking device can be collected to determine the size and shape of the locking device. Next, the size and shape can be used alone or together with the image data to determine the probability distribution. Although 3D cameras are more expensive than 2D cameras, 3D cameras can provide more information about the locking device, thus providing higher accuracy when determining the lock type.

[0009] In some embodiments of the present disclosure, obtaining the set of probability distributions based on the set of raw data includes: for a given locking device in the set of locking devices, determining the probability of the lock type to which the given locking device belongs based on the given raw data associated with the given locking device in the set of raw data; and obtaining the probability distribution in the set of probability distributions based on the determined probability. Here, each locking device can have a corresponding probability distribution, which represents the probability of the lock types to which the locking device can belong. For example, if there are m lock types, the locking device j can belong to any one of lock types 1, 2, ……, and m. At this time, the probability distribution of the locking device j can be expressed as a vector (p j，1 , p j，2 ,..., p j，m ). Through these embodiments, since the set of probability distributions is determined based on the set of raw data of all locking devices, the set of probability distributions can represent a reliable basis for determining the lock type to which the locking device belongs.

[0010] In some embodiments of the present disclosure, determining the probability of the lock type includes: for a given lock type in the lock types, determining the probability of the given lock type through an image recognition process based on the given raw data. In some embodiments of the present disclosure, determining the probability of the given lock type through a machine learning process based on the given raw data. Nowadays, the development of image recognition and other technologies such as machine learning provides an effective solution for type judgment, and at the same time provides the probability of determining whether it is credible. The higher the probability, the more accurate the recognition. Through these embodiments, the probability of the lock type can be provided based on a solid foundation for image recognition and / or machine learning, so the lock type can be determined in an accurate manner.

[0011] In some embodiments of the present disclosure, the method further includes selecting the set of locks from the plurality of locks based on the relative positions between the plurality of locks installed on the object and the object. Generally, the locks used to fix a single container object in a similar position have the same lock type. Through these embodiments, the locks installed in the similar positions of the container can be selected into the same group, so that the locks in the selected group can provide more information about the lock type.

[0012] In some embodiments of the present disclosure, the relative position includes at least one of a corner position and an intermediate position. The container can be fixed with different types of locks. For a long-sized container, the locks at the four corner positions have the same type. For a short-sized container, the total length of the two short-sized containers is equal to the length of the long-sized container, so the two short-sized containers can be connected together in their length direction by different locks to form a long-sized container. Therefore, the lock types of the locks installed at the intermediate positions of the combined container can be of different types. Through these embodiments, by dividing the locks into different groups based on the corner positions and the intermediate positions, the locks in a single group can have the same lock type, thereby improving the accuracy of lock type determination.

[0013] In some embodiments of the present disclosure, determining the lock type of the lock includes: generating a probability list based on a comparison between the value associated with the target lock type in the set of probability distributions and the values associated with the plurality of lock types in the set of probability distributions; and determining the lock type of the lock based on the generated probability list. Through these embodiments, the raw data of all the locks in the same group can be considered when generating the probability list, so the probability list can reflect the probabilities of all the lock types to which the lock may belong.

[0014] In some embodiments of the present disclosure, determining the lock type of a lock based on the generated probability list includes: in response to the highest probability in the probability list being higher than a predefined threshold, identifying the lock type of the lock as the lock type corresponding to the highest probability. Through these embodiments, a threshold can be determined in advance to represent a reliable criterion. If the highest probability is higher than the threshold, it indicates that the lock type corresponding to the highest probability is reliable and acceptable; otherwise, the lock type corresponding to the highest probability can be discarded. In these embodiments, only reliable lock types can be output to further control the robot system to remove the lock.

[0015] In some embodiments of the present disclosure, the method further includes: in response to the maximum probability in the probability list being lower than a predefined threshold, providing an alarm for indicating a potential error. Through these embodiments, outputting unreliable lock types can be prevented. In this case, the above method can be restarted for another round of lock type determination until a reliable lock type is determined.

[0016] In a second aspect, example embodiments of the present disclosure provide an apparatus for determining the lock type of a lock. The apparatus includes: a collection unit configured to collect a set of raw data of a set of locks from a set of sensors respectively, the set of locks being installed on an object at a set of positions respectively, and the set of locks belonging to at least one lock type among multiple lock types; an acquisition unit configured to obtain a set of probability distributions based on the set of raw data, where the probability distributions in the set of probability distributions are associated with the probabilities of the locks in the set of locks and the lock types to which the locks belong; and a determination unit configured to determine the lock type of the lock based on the set of probability distributions.

[0017] In some embodiments of the present disclosure, the acquisition unit is further configured to: for a given lock in the set of locks, determine the probability of the lock type to which the given lock belongs based on the given raw data associated with the given lock in the set of raw data; and obtain the probability distribution in the set of probability distributions based on the determined probability.

[0018] In some embodiments of the present disclosure, the acquisition unit is further configured to: for a given lock type in the lock types, determine the probability of the given lock type based on any one of the following: an image recognition process based on the given raw data; and a machine learning process based on the given raw data.

[0019] In some embodiments of the present disclosure, the determination unit is further configured to: generate a probability list based on a comparison between the value associated with the target lock type in the set of probability distributions and the values associated with multiple lock types in the set of probability distributions; and determine the lock type of the lock based on the generated probability list.

[0020] In some embodiments of the present disclosure, the determining unit is further configured to: in response to the highest probability in the probability list being higher than a predefined threshold, identify the lock type of the lock as the lock type corresponding to the highest probability; and in response to the highest probability in the probability list being lower than the predefined threshold, provide an alert for indicating a potential error.

[0021] In some embodiments of the present disclosure, the apparatus further includes: a selection unit configured to select the set of locks from the plurality of locks based on the relative positions between the plurality of locks installed on the object and the object.

[0022] In some embodiments of the present disclosure, the relative position includes at least one of a corner position and an intermediate position.

[0023] In some embodiments of the present disclosure, the lock includes a twist lock and the object includes a container, and the apparatus further includes: an indicating unit configured to indicate to the robotic system to remove the lock from the object based on the type of the lock.

[0024] In some embodiments of the present disclosure, the sensors in the set of sensors include a dimension measurement camera, and the raw data includes dimension data of the lock.

[0025] In some embodiments of the present disclosure, the sensors in the set of sensors include an image measurement camera, and the raw data includes image data of the lock.

[0026] In a third aspect, example embodiments of the present disclosure provide a computer system for determining the lock type of a lock. The computer system includes: a computer processor coupled to a computer-readable memory unit, the memory unit including instructions that, when executed by the computer processor, implement the method for determining the lock type of a lock according to the first aspect of the present disclosure.

[0027] In a fourth aspect, example embodiments of the present disclosure provide a computer-readable medium having instructions stored thereon that, when executed on at least one processor, cause the at least one processor to execute the method for determining the lock type of a lock according to the first aspect of the present disclosure.

[0028] In a fifth aspect, example embodiments of the present disclosure provide a system for determining the lock type of a lock. The system includes: a set of sensors for collecting a set of raw data of a set of locks; and a computer system according to the third aspect of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 A schematic diagram showing a working environment in which embodiments of the present disclosure can be implemented;

[0030] Figure 2 A schematic diagram for determining a lock type according to an embodiment of the present disclosure;

[0031] Figure 3 A flowchart showing a method for determining a lock type according to an embodiment of the present disclosure;

[0032] Figure 4 A schematic diagram showing a method for obtaining the dimensions of a lock according to an embodiment of the present disclosure;

[0033] Figure 5 A schematic diagram showing a method for dividing multiple locks into groups according to an embodiment of the present disclosure;

[0034] Figure 6 A schematic diagram showing a data structure of a probability distribution according to an embodiment of the present disclosure;

[0035] Figure 7 A schematic diagram showing a method for generating a probability list according to an embodiment of the present disclosure;

[0036] Figure 8 A schematic diagram showing a device for determining a lock type according to an embodiment of the present disclosure;

[0037] Figure 9 A schematic diagram showing a computer system for determining a lock type according to an embodiment of the present disclosure; and

[0038] Figure 10 A schematic diagram showing a system for determining a lock type according to an embodiment of the present disclosure.

[0039] Throughout the drawings, the same or similar reference numerals are used to denote the same or similar elements. Detailed Description of the Embodiments

[0040] Now, the principles of the present disclosure will be described with reference to several exemplary embodiments shown in the drawings. Although the exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the description of these embodiments is only for facilitating those skilled in the art to better understand and thus implement the present disclosure, rather than limiting the scope of the present disclosure in any way.

[0041] For the sake of description, reference will be made to Figure 1 to provide a general description of the environment of the embodiments. Figure 1 A schematic diagram of a working environment 100 in which embodiments of the present disclosure can be implemented is shown. In the context of the present disclosure, embodiments will be described by taking a container as an object on which multiple locks are installed as an example. Specifically, in Figure 1 , the object 110 (e.g., a container) is unloaded from a ship, and the locks 120 and 122 are still installed on the container. First, the lock types of the locks 120 and 122 can be determined, and then the locks 120 and 122 can be removed by the robot system 130 based on the determined lock types.

[0042] Currently, human workers check the locks one by one and determine the lock types. On the one hand, the cost of hiring labor is very high; on the other hand, even experienced workers may make mistakes due to the deformation and corrosion of the locks during long-term use. Therefore, it is desired to determine the lock types of the locks in a more effective and convenient way.

[0043] Reference will be made to Figure 2 to understand more details about how to determine the lock types of the locks. Figure 2 FIG. 200 shows a schematic diagram for determining lock types according to an embodiment of the present disclosure. In Figure 2 , a set of sensors 210 and 212 can be equipped to collect the raw data of the locks 120 and 122 respectively. For example, the sensor 210 can collect the raw data 220, and the sensor 212 can collect the raw data 222. The raw data here can be in various formats, such as image data, point cloud data, or a combination thereof. The locks 120 and 122 can be installed on the object 110 at a set of positions respectively. Here, the set of locks can belong to at least one of multiple lock types.

[0044] In addition, a set of probability distributions can be obtained based on the set of raw data. In Figure 2 , the probability distribution 230 is obtained from the raw data 220 respectively, and the probability distribution 232 is obtained from the raw data 222. Here, the probability distribution 230 is associated with the probability of the lock 120 and the lock type to which the lock 120 belongs. The probability distribution 232 is associated with the probability of the lock 122 and the lock type to which the lock 122 belongs. Next, the lock type 240 of the locks can be determined based on the probability distributions 230 and 232. Through these embodiments, the lock type 240 of the locks can be determined automatically based on the raw data 220 and 222 without manual intervention. Therefore, compared with manual operation, the performance and accuracy can be greatly improved. Although Figure 2 only two locks are shown in the figure, in another example, the set can include more locks.

[0045] Reference will be made to Figure 3 to provide details of the present disclosure. Figure 3 FIG. 300 shows a flowchart of a method for determining the lock types of the locks according to an embodiment of the present disclosure. In step 310, a set of raw data of a set of locks is obtained from a set of sensors respectively. For example, each sensor can collect the raw data of one lock. For another example, the sensor can move to another position to collect the raw data of another lock. The set of locks can be installed on the object 110 at a set of positions respectively. Referring again to Figure 1 , the lock 120 is installed at the left corner of the object 110, while the lock 122 is installed at the right corner of the object 110.

[0046] In some embodiments of the present disclosure, the locking devices 120 and 122 may include twist locks, and the object 110 may include a container. Here, the locking devices 120 and 122 may be used to connect the container to other containers during transportation. After these containers are unloaded from the ship, the locking devices 120 and 122 should be removed from the container. With these embodiments, the lock type can be automatically determined, and the robotic system 130 can be further instructed to remove the locking devices 120 and 122 from the object 110 based on the type of the locking devices.

[0047] It should be understood that the embodiments of the present disclosure may be very efficient, especially in ports used for unloading containers from ships. Generally, large ships carry thousands of containers, so countless twist locks are required to fix these containers together. With these embodiments, the lock type can be automatically identified, and then the locking devices can be removed without any manual intervention. In addition, the number of human workers can be significantly reduced and the efficiency of the unloading process can be improved.

[0048] Generally, various types of locking devices can be used to connect containers. Here, the set of locking devices may belong to at least one of multiple lock types, and the number of lock types can be predetermined. In one example, the number of lock types can be represented by an integer m, so the lock types can be represented by type 1, type 2, ……, and type m. In addition, the number of locking devices in the set can be represented by an integer n, so the locking devices can be represented by lock 1, lock 2, ……, lock n.

[0049] Since the process of collecting raw data from all sensors is the same, the following paragraphs will provide a detailed process taking sensor 210 as an example. In some embodiments of the present disclosure, sensor 210 may include an image measurement camera, such as a 2D image camera, and the raw data 220 may include the image data of the locking device 120. Referring to Figure 2 , sensor 210 can be a 2D image camera. With these embodiments, the collected image data can be used for further processing to determine the lock type to which the locking device 120 belongs. Nowadays, 2D cameras are inexpensive and widely used for monitoring the unloading process, so these embodiments provide an effective and efficient solution for reusing sensors.

[0050] In addition to 2D cameras, 3D camera devices can also be used to collect raw data. In some embodiments of the present disclosure, sensor 210 may include a dimension measurement camera, such as a 3D image camera, and the raw data 220 may include the dimension data of the locking device 120. With these embodiments, sensor 210 (such as a 3D camera) can be adapted to collect the dimensions of the locking device 120. Specifically, the laser device in the 3D camera can measure the distance between the laser device and almost every point on the surface of the locking device 120. Specifically referring to Figure 4 , Figure 4FIG. 400 shows a schematic diagram for obtaining the dimensions of a lock according to an embodiment of the present disclosure.

[0051] In Figure 4 , the sensor 210 may be equipped with a laser device 410 for measuring the dimensions of the lock 120. During the operation of the sensor 210, the laser device 410 may transmit a signal 420 (such as a laser beam) to the lock 120. The signal 420 may reach a point on the surface of the lock 120, and then the signal 430 may be reflected by the lock 120. The sensor 210 may receive the reflected signal 430 and determine the distance between the laser device 410 and the point on the surface based on the duration between the time point for transmitting the signal 420 and the time point for receiving the signal 430.

[0052] Through the above process, the point cloud data of the lock 120 can be collected to determine the dimensions and shape of the lock 120. The dimension and shape data may be used alone or in combination with the image data to determine the probability distribution 230. Although 3D cameras are more expensive than 2D cameras, 3D cameras can provide more information about the lock 120, which can in turn provide higher accuracy in determining the lock type 240.

[0053] In some embodiments of the present disclosure, the group of locks 120 and 122 may be selected from a plurality of locks installed on the object 110 based on the relative positions between the plurality of locks and the object 110. Generally, the locks used to fix a single container in a similar position belong to the same lock type. Referring again to Figure 1 , the object 110 is a long-sized container, and four locks are used for fixing purposes. Although Figure 1 only two locks 120 and 122 on one side of the container are shown, the other two locks are installed on the hidden side of the container. At this time, the four locks installed at the four corners of the container may be selected. Through these embodiments, the locks installed at similar positions on the container can be selected into the same group, so that the locks in the selected group can provide more information about the lock type.

[0054] In some embodiments of the present disclosure, the relative position includes at least one of a corner position and an intermediate position. Generally, containers can be fixed with different types of locks. For long-sized containers, the locks at the four corner positions have the same type. Regarding short-sized containers, reference will be made to Figure 5 , Figure 5 FIG. 500 shows a schematic diagram for dividing a plurality of locks into groups according to an embodiment of the present disclosure. As Figure 5As shown, the total length of two short-sized containers 510 and 512 is equal to the length of the long-sized container. At this time, the two short-sized containers 510 and 512 can be connected in their length directions by another type of locking device to form a long-sized container.

[0055] In Figure 5 , the locking types of the locking devices 120 and 122 installed at the corner positions of the combined container can be one type, while the locking types of the locking devices 520 and 522 installed at the middle positions of the combined container can be another type. Through these embodiments, by dividing the locking devices into two groups based on the corner positions and the middle positions, the locking devices in a single group can have the same locking type, thereby improving the accuracy of locking type determination.

[0056] Return Figure 3 In block 320 of , a set of probability distributions can be determined based on the set of raw data. Here, the probability distributions in the set of probability distributions can be associated with the locking devices in the set of locking devices and the probabilities of the locking types to which the locking devices belong. For example, the probability distribution 230 can be determined based on the raw data 220, and the probability distribution 232 can be determined based on the raw data 222. Specifically, the probability distribution 230 can be associated with the locking device 120 and the probability of the locking type to which the locking device 120 belongs. More details about the probability distributions will be presented below.

[0057] In some embodiments of the present disclosure, for a given locking device in the set of locking devices, the probability of the locking type to which the given locking device belongs can be determined based on the given raw data in the set of raw data that is associated with the given locking device in the set. Then, based on the determined probability, the probability distribution in the set of probability distributions can be obtained. Here, each locking device can have a corresponding probability distribution, and the probability distribution represents the probability of the locking type to which the locking device may belong.

[0058] Figure 6 FIG. 600 shows a schematic diagram of the data structure of the probability distribution 230 according to an embodiment of the present disclosure. As mentioned in the previous paragraph, the locking device 120 can belong to any one of multiple locking types: type 1, type 2,..., and type m (represented by reference numerals 610, 612,..., and 614). The probability distribution 230 is represented as a vector (p 1，1 , p 1，2 ,..., p 1，m ), where each value in the vector can indicate the probability associated with a locking type. For example, the first value p 1，1 can represent the probability that the locking device 120 belongs to type 1.

[0059] Generally, the probability distribution of the locking device j can be represented in a more common way as (p j，1 , p j，2 ,..., pj，m ) where the value p j，1 can represent the probability that the lock j belongs to type 1, and the value p j，2 can represent the probability that the lock j belongs to type 2,......, and the value p j，m can represent the probability that the lock j belongs to type m. Through these embodiments, this set of probability distributions can provide a reliable basis for determining the lock type to which the lock belongs.

[0060] In some embodiments of the present disclosure, for a given lock type in the lock type, the probability of the given lock type can be determined based on an image recognition process based on given raw data. For example, the image data of the lock 120 can be analyzed to identify the content of the image. The image recognition process can indicate whether the lock belongs to the given lock type with a probability between 0 and 1. For example, if the raw data 220 of the input lock 120 is input, the image recognition process can output the probabilities of type 1, type 2,......, and type m as: (p 1，1 , p 1，2 ,..., p 1，m ). Therefore, the probability distribution 230 can be expressed as (p 1，1 , p 1，2 ,..., p 1，m ).

[0061] Generally speaking, if the raw data of the input lock j is input, the image recognition process can output a probability distribution (p j，1 , p j，2 ,..., p j，m ). Continuing with the above example of including four locks in the group, the image recognition process can provide a set of probability distributions (p j，1 , p j，2 ,..., p j，m ), where j = 1 to 4. Various techniques for image recognition have been developed, and the details will be omitted hereinafter.

[0062] In some embodiments of the present disclosure, the probability of a given lock type can be determined based on a machine learning process using given raw data. Nowadays, machine learning techniques play an important role in image processing, so the probability of the lock type can be output from a trained machine learning model. Various techniques have been developed to train machine learning models, and the details will be omitted hereinafter.

[0063] Referring again to Figure 7, at block 330, the lock type of the locks in the set of locks is determined based on the set of probability distributions. Through these embodiments, the lock type of the locks can be determined automatically based on the raw data collected from a set of sensors without manual intervention, so the performance and accuracy can be greatly improved compared with human operations. Additionally, since container objects are usually locked using the same type of locks, determining the lock type based on a set of raw data of a set of locks can eliminate potential errors caused by deformation and erosion of the locks.

[0064] In some embodiments of the present disclosure, to determine the lock type of the locks, a probability list can be generated based on a comparison between the values associated with the target lock type in the set of probability distributions and the values associated with multiple lock types in the set of probability distributions. Regarding generating the probability list, reference will be made to Figure 7 .

[0065] Figure 7 FIG. 700 shows a schematic diagram for generating a probability list 730 according to an embodiment of the present disclosure. In Figure 7 , a set of probability distributions 230, 710,......, and 712 can be determined according to the steps described in block 320. Additionally, these probability distributions 230, 710,......, and 712 can be used to determine the members 720, 722,......, and 724 in the probability list 730. In one example, the values associated with the target lock type in the set of probability distributions can be compared with the values associated with multiple lock types in the set of probability distributions. Here, the target lock type can be any lock type, and solutions such as Bayes' theorem can be used for the comparison.

[0066] In one example, to determine the member associated with lock type i, the following formula 1 can be used:

[0067]

[0068] In formula 1, P i represents the probability that the set of locks belongs to type i, m represents the number of lock types, n represents the number of locks in the set, and p j，i represents the probability that lock j belongs to type i. The numerator in the above formula is related to the value associated with the target lock type i in the set of probability distributions, and the denominator is related to the values associated with multiple lock types (type 1, type 2,......, and type m) in the set of probability distributions. It should be understood that the above formula 1 is only an example formula for determining the probabilities in the probability list. In other embodiments of the present disclosure, other formulas can be adopted as long as these formulas can reflect the candidate types of the set of locks. For example, the operator Π can be replaced by the operator ∑.

[0069] In addition, the lock type of the lock can be determined based on the generated probability list 730. Specifically, the highest probability can be selected to determine the lock type of the group of locks. In some embodiments of the present disclosure, if the highest probability in the probability list 730 is higher than a predefined threshold, the lock type of the lock can be determined as the lock type corresponding to the highest probability. For example, the probability list 730 can be expressed as (P1, P2,..., P m ). If the first member P1 is the maximum value greater than the threshold, the lock type can be determined as type 1. Again, for example, if the i-th member P i is the maximum value, the lock type can be determined as type i. Here, the threshold can be predetermined to represent a reliability criterion. If the highest probability is higher than the threshold, it indicates that the lock type corresponding to the highest probability is reliable and acceptable, otherwise the lock type corresponding to the highest probability can be discarded. In these embodiments, only reliable lock types can be output to further control the robot system to remove the locks.

[0070] In some embodiments of the present disclosure, if the maximum probability in the probability list 730 is lower than a predefined threshold, an alarm for indicating a potential error can be provided. Through these embodiments, unreliable lock types can be prevented from being output. Next, the above method can be restarted for another round of lock type determination until a reliable lock type is determined.

[0071] In some embodiments of the present disclosure, a device 800 for determining the lock type of a lock is provided. Figure 8 A schematic diagram of a device 800 for determining a lock type according to an embodiment of the present disclosure is shown. As Figure 8 shown, the device 800 may include: a collection unit 810 configured to collect a set of raw data of a set of locks from a set of sensors respectively, the set of locks being installed at a set of positions on an object respectively, and the set of locks belonging to at least one lock type among multiple lock types; an acquisition unit 820 configured to obtain a set of probability distributions based on the set of raw data, where the probability distributions in the set of probability distributions are associated with the probability of the locks in the set of locks and the lock types to which the locks belong; and a determination unit 830 configured to determine the lock type of the lock based on the set of probability distributions.

[0072] In some embodiments of the present disclosure, the acquisition unit 820 is further configured to: for a given lock in the set of locks, determine the probability of the lock type to which the given lock belongs based on the given raw data associated with the given lock in the set of raw data; and based on the determined probability, obtain the probability distribution in the set of probability distributions.

[0073] In some embodiments of the present disclosure, the acquisition unit 820 is further configured to: for a given lock type in the lock type, determine the probability of the given lock type based on any one of the following: an image recognition process based on the given raw data; and a machine learning process based on the given raw data.

[0074] In some embodiments of the present disclosure, the determining unit 830 is further configured to: generate a probability list based on a comparison between a value associated with a target lock type in the set of probability distributions and values associated with a plurality of lock types in the set of probability distributions; and determine the lock type of the lock based on the generated probability list.

[0075] In some embodiments of the present disclosure, the determining unit 830 is further configured to: identify the lock type of the lock as the lock type corresponding to the highest probability in response to the highest probability in the probability list being higher than a predefined threshold; and provide an alert for indicating a potential error in response to the highest probability in the probability list being lower than the predefined threshold.

[0076] In some embodiments of the present disclosure, the apparatus 800 further includes: a selection unit configured to select the set of locks from a plurality of locks based on a relative position between the plurality of locks installed on an object and the object.

[0077] In some embodiments of the present disclosure, the relative position includes at least one of a corner position and an intermediate position.

[0078] In some embodiments of the present disclosure, the lock includes a twist lock and the object includes a container, and the apparatus 830 further includes: an indicating unit configured to indicate to a robot system to remove the lock from the object based on the type of the lock.

[0079] In some embodiments of the present disclosure, the sensors in the set of sensors include a dimension measurement camera, and the raw data includes dimension data of the lock.

[0080] In some embodiments of the present disclosure, the sensors in the set of sensors include an image measurement camera, and the raw data includes image data of the lock.

[0081] In some embodiments of the present disclosure, a system 900 for determining a lock type is provided. Figure 9 A schematic diagram of a computer system 900 for determining a lock type according to an embodiment of the present disclosure is shown. As Figure 9 shown, the system 900 may include a computer processor 910 coupled to a computer-readable memory unit 920, and the memory unit 920 includes instructions 922. When executed by the computer processor 910, the instructions 922 may implement the method for determining a lock type described in the foregoing paragraphs, which will not be elaborated hereinafter.

[0082] In some embodiments of the present disclosure, a computer-readable medium for determining the lock type of a lock is provided. Instructions are stored on the computer-readable medium, and when executed on at least one processor, the instructions may cause the at least one processor to execute the method for determining the lock type of a lock described in the foregoing paragraphs, which will not be elaborated hereinafter.

[0083] In some embodiments of the present disclosure, a system for determining the lock type of a lock is provided. Figure 10 A schematic diagram of a system 1000 for determining the lock type according to an embodiment of the present disclosure is shown. The system 1000 includes: a set of sensors 210, …, 212 for collecting a set of raw data of a set of locks 120, …, 122; and a computing system 1010 for determining the lock type of the locks in the set of locks according to the present disclosure.

[0084] Generally, various embodiments of the present disclosure can be implemented using hardware or dedicated circuits, software, logic, or any combination thereof. Some aspects can be implemented using hardware, while other aspects can be implemented using firmware or software that can be executed by a controller, a microprocessor, or other computing devices. Although various aspects of the embodiments of the present disclosure are illustrated and described as block diagrams, flowcharts, or using some other graphical representation, it should be understood that, by way of non-limiting example, the blocks, devices, systems, techniques, or methods described herein can be implemented using hardware, software, firmware, dedicated circuits or logic, general hardware or a controller or other computing devices, or some combination thereof.

[0085] The present disclosure also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as the instructions included in program modules, which are executed in a device on a target real or virtual processor to perform the processes or methods described above with reference to Figure 3 described. Generally, program modules include routines, programs, libraries, objects, classes, components, data structures, etc. that perform specific tasks or implement specific abstract data types. In various embodiments, the functions of the program modules can be combined or split as needed among the program modules. The machine-executable instructions of the program modules can be executed within a local or distributed device. In a distributed device, the program modules can be located in both local and remote storage media.

[0086] The program code for performing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or a controller of a general-purpose computer, a special-purpose computer, or other programmable data processing devices, such that when the program codes are executed by the processor or the controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0087] The above program code can be embodied on a machine-readable medium, which can be any tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include an electrical connection having one or more wires, a portable computer floppy disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0088] Moreover, although operations are described in a particular order, this should not be construed as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed to obtain the desired result. In some cases, multitasking and parallel processing may be advantageous. Also, although several specific implementation details are included in the above discussion, these should not be construed as limitations on the scope of the present disclosure, but rather as descriptions of features that may be specific to particular embodiments. Certain features described in the context of separate embodiments may also be implemented in combination in a single embodiment. On the other hand, the various features described in the context of a single embodiment may also be implemented separately or in any suitable sub-combination in multiple embodiments.

[0089] Although the subject matter has been described in language specific to structural features and / or methodological acts, it is to be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or acts described above. Rather, the above specific features and acts are disclosed as example forms of implementing the claims.

Claims

1. A method for determining the lock type of a lock, the method comprising: Collecting a set of raw data of a set of locks from a set of sensors respectively, the set of locks being installed on an object at a set of positions respectively, and the set of locks belonging to at least one lock type among a plurality of lock types; Obtaining a set of probability distributions based on the set of raw data, the probability distributions in the set of probability distributions being associated with the probability of the locks in the set of locks and the lock types to which the locks belong; And Determining the lock type of the lock based on the set of probability distributions, Wherein determining the lock type of the lock comprises: Generating a probability list based on a comparison between the values associated with a target lock type in the set of probability distributions and the values associated with the plurality of lock types in the set of probability distributions; And Determining the lock type of the lock based on the generated probability list.

2. The method according to claim 1, wherein obtaining the set of probability distributions based on the set of original data includes: Regarding a given lock in the set of locks, Determining the probability of the lock type to which the given lock belongs based on the given raw data associated with the given lock in the set of raw data; and Obtaining the probability distribution in the set of probability distributions based on the determined probability.

3. The method according to claim 2, wherein determining the probability of the lock type to which the given lock belongs includes: Regarding a given lock type in the lock types, determining the probability of the given lock belonging to the given lock type based on any one of the following: An image recognition process of the given raw data; and A machine learning process of the given raw data.

4. The method according to claim 1, wherein determining the lock type of the lock based on the generated probability list comprises: In response to the highest probability in the probability list being higher than a predefined threshold, identifying the lock type of the lock as the lock type corresponding to the highest probability; And The method further comprises: in response to the highest probability in the probability list being lower than a predefined threshold, providing an alarm for indicating a potential error.

5. The method according to claim 1, further comprising: Selecting the set of locks from the plurality of locks based on the relative positions between the plurality of locks installed on the object and the object.

6. The method according to claim 5, wherein the relative position includes at least one of a corner position and an intermediate position.

7. The method according to claim 1, wherein the lock includes a twist lock and the object includes a container, and the method further comprises: Based on the type of the lock, instructing a robotic system to remove the lock from the object.

8. The method according to claim 1, wherein the sensors in the set of sensors include dimension measurement cameras, and the raw data includes dimension data of the locks.

9. The method according to claim 1, wherein the sensors in the set of sensors include image measurement cameras, and the raw data includes image data of the locks.

10. A device for determining the lock type of a lock, the device comprising: A collection unit configured to collect a set of raw data of a set of locks from a set of sensors respectively, the set of locks being installed on an object at a set of positions respectively, and the set of locks belonging to at least one lock type among a plurality of lock types; An acquisition unit, configured to acquire a set of probability distributions based on the set of original data, where the probability distributions in the set of probability distributions are associated with the probability of a lock in the set of locks and the lock type to which the lock belongs; And A determination unit, configured to determine the lock type of the lock based on the set of probability distributions, Wherein the determination unit is further configured to: Generate a probability list based on a comparison between the values associated with the target lock type in the set of probability distributions and the values associated with the multiple lock types in the set of probability distributions; And Determine the lock type of the lock based on the generated probability list.

11. The apparatus according to claim 10, wherein the acquisition unit is further configured to: for a given lock in the set of locks, Determine the probability of the lock type to which the given lock belongs based on the given original data associated with the given lock in the set of original data; and Acquire the probability distribution in the set of probability distributions based on the determined probability.

12. The apparatus according to claim 11, wherein the acquisition unit is further configured to: for a given lock type in the lock types, determine the probability of the given lock belonging to the given lock type based on any one of the following: An image recognition process of the given original data; and A machine learning process of the given original data.

13. The apparatus according to claim 10, wherein the determination unit is further configured to: In response to the highest probability in the probability list being higher than a predefined threshold, identify the lock type of the lock as the lock type corresponding to the highest probability; and In response to the highest probability in the probability list being lower than a predefined threshold, provide an alarm for indicating a potential error.

14. The apparatus according to claim 10, further comprising: A selection unit, configured to select the set of locks from the multiple locks based on the relative position between the multiple locks installed on the object and the object.

15. The apparatus according to claim 14, wherein the relative position includes at least one of a corner position and an intermediate position.

16. The apparatus according to claim 10, wherein the lock includes a twist lock and the object includes a container, and the apparatus further comprises: An indication unit, configured to indicate to the robotic system to remove the lock from the object based on the type of the lock.

17. The apparatus according to claim 10, wherein the sensor in the set of sensors includes a dimension measurement camera, and the original data includes the dimension data of the lock.

18. The apparatus according to claim 10, wherein the sensor in the set of sensors includes an image measurement camera, and the original data includes the image data of the lock.

19. A computer system for determining the lock type of a lock, comprising: A computer processor, coupled to a computer-readable memory unit, the memory unit including instructions that, when executed by the computer processor, implement the method according to any one of claims 1 to 9.

20. A computer-readable medium storing instructions that, when executed on at least one processor, cause the at least one processor to perform the method according to any one of claims 1 to 9.

21. A system for determining the lock type of a lock, comprising: a set of sensors for collecting a set of raw data of a set of locks; and the computer system according to claim 19, for determining the lock type of the locks in the set of locks.

Images