Patent Application
20240339044
The present invention is in the field of ergonomic training. It may advantageously be used in the context of order picking.
EP 3 454 744 A1 shows systems and devices for motion tracking, assessment, and monitoring and methods of use thereof.
US 2021/0290176 A1 shows systems and methods for improving workplace safety.
US 2021/0389817 A1 shows methods and apparatuses for actions, activities and task classifications based on machine learning techniques.
It is an object of the present invention to provide for at least one of the following aspects:
This is achieved by a system for ergonomic training as defined in claim 1. It is also achieved by a method according to the co-ordinate claim. Advantageous embodiments are shown in the dependent claims and in the following description and the figures.
A system for ergonomic training, comprises a server.
The system furthermore comprises a plurality of stations, each station including at least one predefined object to be handled by a test person, wherein a predefined manner of handling the predefined object is defined for each predefined object. Moreover, each station comprises a localization arrangement. Each station may comprise one or more localization arrangements, which may for instance be provided on one or more of the predefined objects and/or separate from the predefined objects.
The system furthermore comprises a mobile kit to be carried by a test person. The mobile kit includes at least three wearable inertial sensors to be worn by the test person and configured for recording and storing sensor data indicative of a movement of the test person. The mobile kit further includes a localization device configured for interaction with the localization arrangement(s) of each station.
The system is configured to determine that the localization device is in proximity of the localization arrangement of a specific station, based on an interaction between the localization device and the localization arrangement of the specific station.
The system is furthermore configured for storing information enabling identification of the specific station, in response to determining the proximity between the localization arrangement of the specific station and the localization device, and for storing subsequently recorded sensor data in a manner associable with the information enabling identification of the specific station.
The server is configured for receiving the information enabling identification of the specific station and the sensor data associable with it.
The server is configured for retrieving information on the at least one predefined object and on the predefined manner of handling the at least one predefined object for the specific station.
The server comprises a machine learning algorithm configured to process the sensor data, along with the information on the at least one predefined object and on the predefined manner of handling the at least one predefined object, in order to evaluate a quality of movements carried out by the test person when handling the at least one predefined object in the predefined manner.
A method for ergonomic training according to the present application may comprise the following steps:
A step of providing a plurality of stations, each station including at least one predefined object to be handled by a test person, wherein a predefined manner of handling the predefined object is defined for each predefined object, and each station comprising a localization arrangement.
A step of equipping a test person with a mobile kit, the mobile kit including at least three wearable inertial sensors to be worn by the test person and con-figured for recording and storing sensor data indicative of a movement of the test person, the mobile kit further including a localization device configured for interaction with the localization arrangement of each station.
A step of providing instructions to the test person to handle at least a subset of the at least one predefined object of at least a subset of the plurality of stations in the respective predefined manner.
Within the method, test person may move consecutively to each of the stations of the subset of the plurality of stations. There, the test person handles the subset of the at least one predefined object at each of these stations in the respective predefined manner.
Moreover, the following further steps may be executed:
An interaction between the localization device and the localization arrangement of a specific station is detected. Based on the detected interaction, it is determined that the localization device is in proximity of the localization arrangement of a specific station.
Information enabling identification of the specific station is stored, in response to determining the proximity between the localization arrangement of the specific station and the localization device. Subsequently, sensor data is recorded in a manner associable with the information enabling identification of the specific station.
The information enabling identification of the specific station and the sensor data associable with it are uploaded to a server.
Information on each specific object pertaining to the subset of the at least one predefined object is retrieved. Additionally, information on the predefined manner of handling each specific object is retrieved.
A machine learning algorithm is employed for processing the sensor data, along with the information on each specific object, and along with the information on the predefined manner of handling each specific object, in order to evaluate a quality of movements carried out by the test person when handling each specific object in the predefined manner.
The machine learning algorithm may in particular include an algorithm for human activity recognition. This human activity recognition may be employed in order to identify a specific human activity when the object is being handled. Based on this identification, the quality of the movements may be evaluated in a particularly targeted way. This will be explained further below.
According to the present application, a method may include using the system shown and described herein. In particular, the above-described method may employ some or all of the system components shown and described herein. It should be noted that features described herein with respect to the system may also be claimed in conjunction with the method. Vice versa, aspects explained in the context of the method may also be claimed in conjunction with the device. In particular, components of the system may be specifically configured to facilitate and/or carry out steps or aspects of the method.
The device and method enable a versatile assessment of ergonomics. Therein, they may be dynamically employed and adjusted for use in a specific environment found at a client's or user's location or facilities. The test person can then complete a course that can be individually set up, and is not subject to any constraints in terms of size or spatial extent. The environment does typically not need to be altered, and there is no need to install cameras or other surveillance equipment.
While generally not being limited in size, the system may for instance comprises at least 5 stations or at least 20 stations. For instance, it may comprise 30 stations.
At each of the plurality of stations, for example, at least two or at least three predefined objects may be provided, and among those, any subset may be handled by the test person. The various objects may be identical to each other in terms of size and/or weight, or they may differ from each other in terms of size and/or weight.
The subset may be chosen based on a specific aim of the test to be performed and for instance in accordance with the test person's work description. For example, the test person may perform 2-3 picks per station, performing different kinds of movements with different kinds of objects.
If there are several predefined objects at a given station, they may be handled in the same manner, or in a manner different from each other. For example, the type of object and their size and shape may be the same for all objects within one station, and they may be different for objects included in another station. In another example, the objects may differ from each other within a single station.
These different ways and combinations of setting up the system and carrying out the method provide flexibility in which types of movements can be assessed and in how the system may be integrated into a given location or environment.
In an example, the various predefined objects may include predefined objects that differ from each other in weight and/or size. Then, the server may be configured for retrieving information on weight and/or size of each predefined object and for including this information in the evaluation of the quality of movements carried out by the test person when handling the respective predefined object in the predefined manner.
The device and method advantageously facilitate taking into account characteristics of the predefined objects in order to evaluate the movements, since the information of the predefined objects is retrieved and may be used when assessing whether or not handling of the object has been performed in an ergonomically advantageous way. For example, the information on the predefined object may include information on size and/or weight. By taking these aspects into account, an unergonomic or harmful movement performed with a heavier object may be assigned a stronger negative rating than an unergonomic or harmful movement performed with a light object. For example, a heavier object may require a different posture than a light object, and this may readily be taken into account here. Also, problematic postures can be evaluated according to whether they are due to the dimensions of an unmanageable object or whether they are due to unnecessary and therefore avoidable movements of the test person.
For example, the predefined manner of handling the predefined object may include, at least some of the stations, one or more of the following actions: picking the object, lifting up the object, removing the object from a shelf, putting down the object, inserting the object into a shelf, carrying the object over a distance, moving the object by way of a hand cart. For example, inserting the object into a shelf may include inserting the object into the shelf at a predefined height, and/or inserting several objects at various different heights.
For example, evaluation of the quality of the movements may include identification of one or more of the following actions performed by the test person, based on the sensor data: walking, standing, driving a vehicle, handling a hand cart, handling an object using one hand, handling an object using two hands, lifting an object, holding an object, carrying an object, bending a back of the test person, torsion of the back of the test person, handling an object in front of a body of the test person, handling an object above a shoulder of the test person. These actions may be identified by way of the human activity recognition of the machine learning algorithm. Subsequently, the quality of these activities and of the movements performed during these activities can be evaluated in an advantageous manner.
To this end, the system may be configured to determine classes of movement associated with the above-listed actions. The at least three wearable inertial sensors gather relevant information for enabling identification of these classes of movement. The server employs machine learning to determine classes of movement and/or posture, based on the received sensor data. From this, the above-listed actions may for instance be derived and their quality and ergonomicity can subsequently be determined.
The at least three wearable inertial sensors include a first wearable inertial sensor that is included in a waist- or hip-belt. Additionally and/or alternatively, the at least three wearable inertial sensors may comprise a second wearable inertial sensor that is included in a wrist band and, for example, a third wearable inertial sensor that is included in a further wrist band. The test person may carry the belt with the first inertial sensor around their waist or hip and may carry the wrist bands with the second and third inertial sensors around their wrists, i.e., one at the left wrist and one at the right wrist.
In an embodiment, the system comprises, for each test person, precisely the three above-mentioned types of wearable inertial sensors, i.e., the first wearable inertial sensor that is included in the waist- or hip-belt, the second wearable inertial sensor that is included in the wrist band, and the third wearable inertial sensor that is included in the further wrist band, and no further sensors are envisioned for a given test person. Of course, several sets of sensors may be provided, so that more than one test person can employ the system at the same time. It was found that provision of these three sensors represents an advantageous setup that enables capturing motions as required for the method and system, while keeping the system and data processing required comparably slim and simple. In particular, this setup enables human activity recognition in an advantageous manner.
In an example, the wearable inertial sensors include an accelerometer and/or a gyroscope and/or a magnetometer. For example, each of the three wearable inertial sensors may include one or more of an accelerometer and/or a gyroscope and/or a magnetometer.
For example, a recording rate of the sensors may be at least 33 Hz. In another example, the recording rate of the sensors is at least 50 Hz. For example, the recording rate may be 100 Hz.
In an example of carrying out the system or method, the mobile kit may comprise a mobile electronic device (MED). The MED may be carried out as a Portable Data Terminal (PDT) and/or a smartphone and/or a portable computer.
In an example, the mobile electronic device may include the localization device of the mobile kit. However, the localization device may for instance also be included in the wearables of the wearable inertial sensors. The localization device may also be provided separate from the inertial sensors and a possible mobile electronic device.
In an example, the mobile electronic device may comprise an output device for providing instructions to the test person.
In an example, the localization arrangement(s) of the station(s) may include a first wireless communication device, and the localization device of the mobile kit may comprise a second wireless communication device, configured for interaction with the first wireless communication device. The first and/or second wireless communication devices may for instance be carried out as WiFi-devices, Ultra-Wideband-Devices, Radio-Wave-Devices. In an example, the first wireless communication device may provide a beacon, such as a Bluetooth beacon.
Additionally or alternatively, the localization arrangement may comprise a scannable code, such as a barcode or a QR-Code, and the localization device may comprise a scanner for scanning the scannable code. Proximity may then be determined, since it is necessary for the test person to be close to the code, in order to scan the code.
Each station may comprise one or more localization arrangements. They may be provided on one or more of the objects of a given station, for example. They may however also be provided separate from the objects. In an example, a scannable code may be provided on each of the objects. This may facilitate providing instructions relating to a specific scanned object to the test person and/or retrieving information on a given object during the evaluation carried out on the server.
In an example of the system or method, the mobile kit may include at least one memory for storing the information enabling identification of the specific station and the sensor data associable with it. For example, the at least one memory may include a memory in some or in each one of the at least three wearable sensors. For example, a separate memory may be provided for each of the wearable inertial sensors, or a joint memory may be provided, communicatively connected to all of the wearable inertial sensors worn by the test person. Additionally or alternatively, the mobile kit may comprise a memory in the mobile electronic device.
The mobile kit may include a communication system, in particular a wireless communication system, configured for communication between the mobile electronic device and the at least three wearable sensors.
This communication system enables interaction between the components of the mobile kit, for instance in order to transfer information on a detected proximity, as detected by the mobile electronic device, to the at least three wearable sensors, in order to enable storing sensor data in a manner associable with the detected proximity.
For example, in an example where the localization device is provided in the mobile electronic device, data relating to its interaction with a localization arrangement of a station may be transferred to the wearable sensors. The wearable sensors may comprise one or more memories, and the data relating to the interaction with a localization arrangement may be saved in these memories along with the sensor data that is recorded in a manner temporally related to the detected interaction with the localization arrangement. The detected interaction with the localization arrangement may enable an identification of a specific station, and of a specific object at that station that is supposed to be handled by the test person at the station, and of a specific manner in which the object is supposed to be handled. This information may then be paired with the sensor data to evaluate quality of movements.
In an example, the system may comprise a docking station for at least some of the components of the mobile kit. The docking station may be connected to the server and enable upload of the information enabling identification of the specific station and the sensor data associable with it from the memory to the server. For example, data upload to the server from or through the docking station may be done using WiFi and/or an internet connection, enabling providing the system away from the server, at any desired location.
For example, the docking station may be configured to receive the wearable sensors. In an example, the memories of the wearable sensors hold the necessary data for evaluating the movements, wherein data from an optional mobile electronic is communicated to the wearable sensors before docking. This way, it is sufficient to dock the wearable inertial sensors, and the mobile electronic device does not need to be docked. Vice versa, it is also possible to have a memory in the mobile electronic device, and transferring the sensor data from the wearable sensors to the mobile electronic device, and then establishing a data transfer from the mobile electronic device to the server.
In another example, it may also be envisioned to establish a direct communication with the server, for the wearable inertial sensors and/or for the optional mobile electronic device, wherein data may be uploaded without use of a docking station.
In possible embodiments of the system and method, it may be envisioned that a supervisor oversees the ergonomic training. Optional ways of carrying out the system include an interaction device for interaction with the supervisor, in particular to enable input by the supervisor. The system may be designed to enable the supervisor to provide an additional assessment or evaluation of the movements carried out by the test person. Additionally or alternatively, the system may be designed to enable the supervisor to design and/or alter a training course.
In an example, the system includes an input device for a supervisor, configured for providing one or more input options to allow the supervisor to input an assessment of a quality of predefined aspects of one or more movements carried out by the test per-on when handling the predefined object in the predefined manner. The server may be configured for receiving the input by the supervisor, and for including said input by the supervisor into the evaluation of the quality of movements carried out by the test person when handling the object in the predefined manner. For example, supervisor input can be provided for each station, or it can be provided jointly for all stations, for example after the test person has terminated all stations. An ergonomic score can be computed that comprises the sensor data and the input from the supervisor, in addition to the information on the predefined objects and the predefined manners of handling the predefined objects.
Input options given to the supervisor may include a requested evaluation of movements, in particular small movements, that are not or not fully captured by the wearable inertial sensors.
In an example of the system or method, the one or more input options for the supervisor include an input option for one or more of:
Input option for the supervisor may be provided for assessing a combined movement of the above-described body parts, such as a combined movement of a combination of at least two of a shoulder, a back, a leg, an arm, a hand, in order to provide an evaluation of whether these body parts are being kept in an advantageous or in a potentially damaging way with respect to each other.
It may also be envisioned to provide input options for the supervisor to specify environment conditions.
In an example of the system or method, the input device for the supervisor may provide an input option for selecting a subset of the plurality of stations by the supervisor and/or for selecting a subset of predefined objects of a given station by the supervisor.
In an example, the input device for the supervisor is connected to the internet, for instance by WiFi, and the input options for the supervisor are provided in a web-interface.
In an example, the method includes:
It may be envisioned that the method of ergonomic training includes carrying out some or all of the above-mentioned steps several times, such as twice or three times, and then comparing the results for assessing a progress made by the test person. For example, it may be envisioned that, in a first iteration, the test person moves consecutively to each of the stations of the subset of the plurality of stations, and handles the subset of the at least one predefined object at each of these stations in the respective predefined manner, as explained above. Subsequently, evaluation of the quality of movements of the first iteration is performed, as explained above. Subsequently, the test person may receive feedback or coaching, which may be based on the evaluation of the quality of movements provided within the method or by way of the system. After this, in a second iteration, the test person may once again move consecutively to each of the stations of the subset of the plurality of stations, and handle the subset of the at least one predefined object at each of these stations in the respective predefined manner once more. Evaluation of the quality of movement for the second iteration may be performed, and the results may be compared to those of the first iteration, in order to track any progress made. Two, three, or more of these iterations may be envisioned. The iterations can take place in quick succession (e.g. the feedback can take place directly after the first iteration, while the test person keeps the inertial sensors on, and then the second iteration can be carried out) or they can take place at intervals of days, weeks or months (with the test person taking the sensors off in the meantime). The system may be configured to store the relevant information pertaining to the test person in a manner associable with the test person. The system may for instance store information on which stations and objects were chosen for the test person, and it may store the results of the evaluation, for later comparison.
The invention will now be explained in an exemplary manner with reference to the appended figures.
Therein:
The system comprises a plurality of stations 100, 200, 300, each of which includes at least one predefined object 101, 102, 103, 201, 202, 203, 301 to be handled by a test person performing the ergonomic training. A predefined manner of handling the predefined object 101, 102, 103, 201, 202, 203, 301 is defined for each predefined object 101, 102, 103, 201, 202, 203, 301. Within the method, ergonomicity of the movements of the test person when handling the objects is evaluated, and the results of the evaluation may be presented to the test person, so that the test person can improve their movement patterns, posture, and the like, and, for example, overcome bad habits in this regard. By providing several stations with several objects, an abundance of different movements may be monitored and assessed, as will be further explained here below. This may provide a particularly broad and accurate picture of the movements of each test person. The system is particularly advantageous in that it may be used in a flexible and versatile manner for a variety of different stations and movements, and it may be used in different places. Specifically, modifying or altering the stations and their objects can be done essentially without requiring modification of the remaining components of the system. These aspects will become more apparent from the explanations given in conjunction with the further figures.
Each station 100, 200, 300 comprises a localization arrangement 110, 210, 310, which enables detection of whether the test person is in the vicinity of the respective station.
The system furthermore comprises a mobile kit 500 to be carried by the test person. The mobile kit 500 includes at least three wearable inertial sensors 501, 502, 503 to be worn by the test person and configured for recording and storing sensor data indicative of a movement of the test person. To this end, one or memories are included in the at least three wearable inertial sensors.
The mobile kit 500 further includes a localization device 510 configured for interaction with the localization arrangement 110, 210, 310 of each station 100, 200, 300. In the example of
The system is configured to determine that the localization device 510 is in proximity of the localization arrangement 110, 210, 310 of a specific station 100, 200, 300, based on an interaction between the localization device 510 and the localization arrangement 110, 210, 310 of the specific station.
If such an interaction is detected and consequently proximity to a specific station is determined, the mobile electronic device interacts with the wearable inertial sensors, to trigger recording of the sensor data, and to submit information on the specific station to the wearable inertial sensors, which are configured to store this information in their memory or memories. To this end, the mobile kit 500 comprises a communication system for communication between the mobile electronic device and the wearable inertial sensors. Consequently, the system is configured for storing information enabling identification of the specific station 100, 200, 300, in response to determining the proximity between the localization arrangement 110, 210, 310 of the specific station and the localization device 510, and for storing subsequently recorded sensor data in a manner associable with the information enabling identification of the specific station 100, 200, 300. While the test person is performing the training, all the relevant sensor data and associated information enabling identification of the station at which the sensor data was recorded, may be kept in the memory or memories of the inertial sensors.
Evaluation of the movements carried out by the test person is subsequently performed on a server 1000, which comprises a machine learning algorithm.
The server 1000 is configured for receiving the information enabling identification of the specific station 100, 200, 300 and the sensor data associable with it. Here, this is done via a docking station 600 where the inertial sensors may be docked after all stations of the training course have been completed by the test person. The docking station 600 is communicatively connected to the server 1000 and the relevant data from the memory or memories of the inertial sensors may be uploaded to the server using WiFi, for example.
The server 1000, on the other hand, receives this information and is configured for retrieving information on the at least one predefined object 101, 102, 201, 202, 301, 302 and on the predefined manner of handling the at least one predefined object 101, 102, 103, 201, 202, 203, 301 for each specific station where an action was performed by the test person.
The server 1000 employs the aforementioned machine learning algorithm in order to process the sensor data, along with the information on the at least one predefined object 101, 102, 103, 201, 202, 203, 301 and on the predefined manner of handling the at least one predefined object 101, 102, 103, 201, 202, 203, 301, in order to evaluate the quality of movements carried out by the test person when handling the at least one predefined object 101, 102, 103, 201, 202, 203, 301 in the predefined manner.
The system in the example of
The server 1000 is configured for receiving the input by the supervisor, and for including said input by the supervisor into the evaluation of the quality of movements carried out by the test person when handling the object 101, 102, 103, 201, 202, 203, 301 in the predefined manner.
The input device 900 for the supervisor furthermore provides an input option for selecting a subset of the plurality of stations 100, 200, 300 by the supervisor and for selecting a subset of predefined objects of each station by the supervisor. The input device 900 for the supervisor is connected to the internet, and the input options for the supervisor are provided in a web-interface.
The station 100 includes the localization arrangement 110, which comprises a scannable code 111 that is exemplarily carried out as a barcode.
The station 100 includes several predefined objects 101, 102, 103 of different weights, sizes and shapes, that are initially provided in a shelf 150. The station 100 additionally includes a hand cart 160.
The mobile electronic device of the mobile kit 500 is here a Portable Data Terminal 550, which includes the localization device 510 in the form of a scanner 511 for scanning the scannable code 111. Moreover, the Portable Data Terminal 550 has an output device 570 in the form of a screen and a wireless communication system 580 for providing instructions to the test person, which may help lead the test person through the course of stations.
When the test person arrives at the station 100, they scan the scannable code 111 using the mobile electronic device. From this, it is clear that the test person is in the vicinity of the specific station 100, because they must be close to the scannable code 111 in order to scan it.
Instructions may be displayed to the test person on the output device 570 of the Portable Data Terminal 550, indicating which of the objects 101, 102, 103 they should manipulate and in which way. Exemplarily, the test person may be instructed to remove object 103 and further objects from the shelf, put them on the hand cart 160 and push the hand cart 160 to a predetermined location. In order to enable provision of such instructions via the portable data terminal, it may be communicatively connected to the server 1000, where the instructions may be stored (and potentially altered by the supervisor), or it may hold these instructions in an internal storage.
Moreover, scanning of the code triggers an interaction between the mobile electronic device 550 and the wearable inertial sensors 501, 502, 503, via the communication system 580, and the wearable inertial sensors may start recording subsequent to this interaction. I.e., in response to the detected interaction, the mobile kit 500 stores the information enabling identification of the specific station and the sensor data associable with it to the memory of the inertial sensors 501, 502, 503.
After the course has been completed, the test person docks the wearable inertial sensors in the docking station 600, which is connected to the server 1000 and enables upload of the information enabling identification of the specific station and the sensor data associable with it from the memories of the wearable inertial sensors 501, 502, 503 to the server 1000.
The localization arrangement 110 of the station includes a first wireless communication device 112, which functions as a Bluetooth beacon.
The mobile kit 500 comprises a mobile electronic that is a smartphone 560. It includes the localization device 510, which comprises a second wireless communication device 512, which is a Bluetooth device configured for interaction with the first wireless communication device 112. The display of the smartphone is used for providing instructions the test person and leading them from one station to the next.
In this case, proximity of the test person to the station 100 may readily be detected via the Bluetooth beacon, and the further step starting the measurements may be initiated based thereon, without requiring the test person perform a scan.
The wearable inertial sensors 501, 502, 503 each include an accelerometer and/or a gyroscope and/or a magnetometer.
Their recording rate of the inertial sensors is 100 Hz.
Turning to
Next to the stacked objects, a shelf 250 is provided. The test person will be required to move at least some of the stacked objects into the shelf. Depending on how many boxes are already in the shelf, the movement of placing the objects will be different. This is also taken into account by the evaluation performed on the server.
When the test person arrives at the station, they will scan the barcode 211 on one of the objects (
It is understood that the features shown and described for stations 100, 200 and 300 may be mixed and matched according to the needs of the specific training.
Within the method, a plurality of stations 100, 200, 300 is provided, each station including at least one predefined object 101, 201, 301 to be handled by a test person, wherein a predefined manner of handling the predefined object is defined for each predefined object, and each station comprising a localization arrangement 110, 210, 310.
A test person is equipped with a mobile kit 500, which includes three wearable inertial sensors 501, 502, 503 to be worn by the test person and which have a memory configured for recording and storing sensor data indicative of a movement of the test person.
The mobile kit 500 further includes a mobile electronic device 550/560 with a localization device configured for interaction with the localization arrangement 110, 210, 310 of each station, and with an output device for providing information to the test person.
An input device 900 is provided to a supervisor.
A server 1000 is provided, which is communicatively connected or connectible to the input device 900, to the mobile electronic device 550/560, and to the inertial sensors 501, 502, 503.
In a first step S1 which is usually performed after taking the preparations above, the supervisor selects, on the input device 900, a subset of the stations 100, 200, 300 and a subset of specific objects 101, 201, 203 of those stations. For each selected object, a manner in which it shall be handled is predefined.
In a next step S2, this selection is passed on to the server 1000, which, in a next step S3, provides this list of stations and objects and manners in which they shall be handled to the mobile electronic device 550/560.
Based thereon, in a subsequent step S5, the mobile electronic device 550/560 then provides instructions to the test person, to move to the specific pre-selected stations and to handle the specific pre-selected predefined objects of these stations in their respective predefined manners. In the example given here, the test person receives an instruction to proceed towards station 100, and to handle object 101 in the predefined manner associated with it.
In a step S4, before or after the instructions are provided to the test subject by way of the mobile electronic device 550/560, the supervisor starts observing the test subject, and starts to take note of his movements.
The test person, having received the above-mentioned instructions, approaches station 100, and, when they arrive at the station 100, in step S6, the localization device of the mobile electronic device 550/560 interacts with the localization arrangement 110 of station 100, which indicates proximity of the test person to station 100. The interaction between the localization device 510 and the localization arrangement is thus detected and based thereon, it can be determined that the localization device 510 and thus the test person is in proximity of the localization arrangement 110 of the specific station 100.
In response to the detected proximity of step S6, step S7 is executed. Therein the mobile electronic device 550/560 communicates with the inertial sensors 501, 502, 503, sending information that is indicative of station 100, and, in a possible example, indicative of the specific object 101, and which signal is saved to the memory of the inertial sensors 501, 502, 503. The inertial sensors 501502, 503 then start recording their sensor data, associated with the received information, which makes it possible to associate the recorded sensor data with station 100 and object 101.
In step S8, the test person proceeds to handle the object 101 of station 100 in the predefined manner, while recording of the sensor data is still active. At the same time, in step 9, the sensors monitor the movements that are being performed by the test person and store the corresponding sensor data in their memory. This sensor data can then be linked to object 101, because the sensors also hold the information indicative of station 100 and object 101 in their memory.
After the test person has finished the assigned tasks at station 100, a further interaction between the localization device of the mobile electronic device 550/560 and the localization arrangement 110 of the station 100 takes place in step S10, and in step S11, the mobile electronic device communicates termination of the station 100 to the inertial sensors, and the sensors store an indication of the termination to their memory and/or stop recording.
Steps S5-S10 are repeated, mutatis mutandis, for the further stations 200, 300 selected by the supervisor.
After the test person has gone through all of the preselected stations and handled all of the preselected objects, the information enabling identification of each specific station and the sensor data associable with it are uploaded to the server 1000, in step S12.
In step 13, input options are given to the supervisor by way of the input device 900, to allow the supervisor to input an assessment of a quality of predefined aspects of one or more movements carried out by the test person when handling the predefined objects in the respective predefined manners.
The input options include input options for one or more of assessing a movement or a posture of a shoulder of the test person, assessing a movement or a posture of a back of the test person, assessing a movement or a posture of a leg or of legs of the test person, assessing a movement or a posture of an arm or of arms of the test person, assessing a movement or a posture of a hand or of hands of the test person.
In step S14, the supervisor's input is transmitted from the input device 900 to the server 1000.
In step S15, the server calculates an ergonomic score based on the information received from the inertial sensors. From the sensor data, the server 1000 identifies for instance that one or more of the following actions have been performed by the test person: walking, standing, driving a vehicle, handling a hand cart, handling an object using one hand, handling an object using two hands, lifting an object, holding an object, carrying an object, bending a back of the test person, torsion of the back of the test person, handling an object in front of a body of the test person, handling an object above a shoulder of the test person. In order to identify these types of movement, the server employs a machine learning algorithm. The server also evaluates whether these actions have been performed in an ergonomically favorable manner, or in a potentially harmful manner.
The server 1000 also retrieves information on each specific object that has been handled by the test person, such as the object's size, shape, and weight. This way, the server can determine whether any of these characteristics are particularly problematic for the test person.
The server also retrieves information on how each object was supposed to be handled. This information may be compared to the movements actually performed by the test person, according to the sensor data, and a potential mismatch may be determined.
Overall, employing a machine learning algorithm, the server processes the sensor data, along with the information on each specific object, and the information on the predefined manner of handling each specific object, in order to evaluate a quality of movements carried out by the test person when handling each specific object in the predefined manner. The server may also include the input given by the supervisor into the evaluation of the quality of movements.
The whole process may be repeated, for example after a training has been completed under the supervision of the supervisor and/or after some weeks or months, giving the test person the opportunity to improve their movements. For example, after the evaluation of the quality of movements, feedback may be provided to the test person, after which, in a second iteration, the test person once again moves consecutively to each of the stations of the subset of the plurality of stations, and handles the subset of the at least one pre-defined object at each of these stations in the respective predefined manner, wherein the quality of movements is subsequently evaluated for the second iteration, in order to track progress made by the test person.
For the scanner 511 and the Bluetooth device 512, there are two different states: Interaction with the localization arrangement of a station/no interaction with any localization arrangement.
The inertial sensors 501, 502, 503 are configured to acquire sensor data indicative of different actions performed by the test person, namely walking, standing, and handling, wherein handling may include driving a vehicle, handling a hand cart, handling an object using one hand, handling an object using two hands, lifting an object, holding an object, carrying an object, bending a back of the test person, torsion of the back of the test person, handling an object in front of a body of the test person, handling an object above a shoulder of the test person. These types of movement are identifiable within the method and system shown herein, based on the acquired sensor data.
The inertial sensors 501, 502, 503 are connected to a memory. The inertial sensors 501, 502, 503, the scanner 511 or Bluetooth device 512 and the memory are connected in such a manner that saving the sensor data to the memory is initiated and terminated based on an interaction of the scanner 511 or Bluetooth device 512 with the localization arrangement of a station. These will now be explained in more detail for the two types of localization devices.
As can be seen from
As can be seen from
The process of
| Number | Date | Country | Kind |
|---|---|---|---|
| 23167105.8 | Apr 2023 | EP | regional |
