DEVICE AND METHOD FOR PREDICTING THE DELIVERY TIME AND ROUTE OF A ROBOT BASED ON AN ESTIMATE OF THE CROWD

Abstract

In one embodiment, predicting a delivery time and route based on a crowd estimate may include defining a plurality of paths to reach a destination, acquiring crowd estimate information for the plurality of paths using sensors on the plurality of paths, calculating traffic volume and latency for each of the plurality of paths based on the crowd estimate information, predicting the delivery time to the destination based on the traffic volume and latency for each of the plurality of paths, and determining the delivery route as the one with the shortest delivery time among the plurality of paths.

Description

TECHNICAL AREA

[0001] The present disclosure relates to a device and a method for predicting a delivery time and a delivery route based on an estimate of the number of people.

[0002] In particular, the present disclosure relates to a device and a method for predicting a delivery time and a delivery route based on an estimate of the number of people in an indoor delivery service using robots. BACKGROUND

[0003] For outdoor deliveries, various variables are used to predict delivery times for unmanned indoor deliveries, such as road conditions and driving behavior. In the case of indoor robot deliveries, the travel time between floors can be predicted fairly accurately by determining elevator stops and load changes. However, this approach fails to account for whether the robot can actually enter the elevator, depending on the level of congestion or overload.

[0004] Particularly in the case that the delivery robot uses the same elevator or the same entrance and exit gate as people, it is possible, depending on the traffic, that the robot will enter the elevator or pass through the entrance and exit gate at the required time. SUMMARY

[0005] The present disclosure seeks to provide, for indoor delivery services using robots, a device and a method for predicting a delivery time and a delivery route based on an estimate of the number of people, which are able to calculate traffic volume and latency by counting the number of people or the number of objects or items on the robot's path of movement using an elevator, an entrance and exit gate, CCTV or the like, and to determine an overcrowding of a particular section, to calculate the delivery time based on the calculated traffic volume and latency, and to determine a path with the shortest delivery time as the delivery route.

[0006] A method for predicting delivery time and route based on crowd estimates may include setting or defining a plurality of routes to reach a destination, collecting or acquiring crowd estimate information for the plurality of routes using sensors on the plurality of routes, calculating traffic volume and latency for each of the plurality of routes based on the crowd estimate information, predicting the delivery time to the destination based on the traffic volume and latency for each of the plurality of routes, and determining the delivery route as the one with the shortest delivery time among the plurality of routes.

[0007] Determining the plurality of paths to reach the destination can involve receiving a stored motion path plan from the robot when a delivery request is received, and determining the plurality of paths based on the motion path plan, where a travel time difference to the destination between paths does not exceed the threshold.

[0008] The sensor can include a camera sensor with CCTV, a lidar sensor, a thermal imaging sensor, an elevator sensor, a robot sensor, and an entry and exit gate sensor.

[0009] The acquisition of information to estimate the number of people on the multitude of paths by sensors on the multitude of paths may include the acquisition of information to count people for each of the multitude of paths by the camera sensor and the thermal imaging sensor.

[0010] The acquisition of information for estimating the crowd size of the multitude of paths by sensors on the multitude of paths may include the acquisition of information for estimating the crowd size of a specific section, including an elevator section and an entrance and exit gate section, on the paths based on entry and exit log data obtained from the elevator sensor and the entrance and exit gate sensor.

[0011] Calculating the traffic volume and latency for each of the multitude of paths based on crowd estimate information may involve calculating the traffic volume based on people count information and the overcrowding of the specific section, and storing the calculated traffic volume in a database.

[0012] Calculating the traffic volume and latency for each of the multitude of paths based on the crowd estimation information may involve calculating the latency based on the time between when the crowd estimation information is collected and when the traffic volume is calculated using that crowd estimation information.

[0013] Predicting the delivery time to the destination based on traffic volume and latency with respect to each of the multitude of routes can involve predicting the delivery time with respect to each of the multitude of routes by adding a required time calculated based on traffic volume and a delay time calculated according to latency.

[0014] The procedure may further include comparing the predicted delivery time and an actual delivery time with respect to the finally determined delivery route after completion of the delivery, and updating a model for predicting the delivery time based on the comparison result if the comparison result meets a certain criterion, and storing the comparison result in a database if the comparison result does not meet the certain criterion.

[0015] If traffic volume and latency cannot be calculated due to a sensor defect or error, predicting the delivery time to the destination based on traffic volume and latency for each of the multitude of routes may also include predicting an average delivery time based on current traffic information, including information on overcrowding and the number of people during a specific current period of time.The duration stored in the database and current information on average latency may include determining the delivery route as the one with the shortest delivery time among the multitude of routes. This may involve generating a multitude of delivery routes and setting priorities based on average delivery time, determining an initial delivery route as the first route with the highest priority, and finally determining the delivery route as the second route with the next highest priority based on traffic volume and latency information updated during the delivery process.

[0016] A device for predicting delivery time and route based on crowd estimates may include a data analyzer configured to acquire crowd estimate information to a destination via a sensor along a multitude of routes; a route analyzer configured to calculate traffic volume and latency to the destination along each of the multitude of routes based on the crowd estimate information; a delivery time prediction unit configured to predict the delivery time to the destination based on the traffic volume and latency along each of the multitude of routes; and a delivery route determination unit configured to determine the delivery route as the one with the shortest delivery time among the multitude of routes.

[0017] When a delivery request is received, the data analyzer can be configured to receive a stored motion path plan from the robot and, based on the motion path plan, determine the multitude of paths where the travel time difference to the destination between the paths does not exceed a threshold.

[0018] The sensor can include a camera sensor with CCTV, a lidar sensor, a thermal imaging sensor, an elevator sensor, a robot sensor, and an entry and exit gate sensor.

[0019] The data analyzer can be set up to collect people counting information for each of the multitude of paths through the camera sensor and the thermal imaging sensor.

[0020] The data analyzer can be configured to detect overcrowding in a specific section, including an elevator section and an entrance and exit gate section, on the pathways based on the input and exit log data received from the elevator sensor and the entrance and exit gate sensor.

[0021] The route analyzer can be set up to calculate traffic volume based on information about the number of people and the overcrowding of the specific section, and to store the calculated traffic volume in a database.

[0022] The path analyzer can be configured to calculate the latency based on the time between when the crowd estimation information is collected and when the traffic volume is calculated using that crowd estimation information.

[0023] The delivery time prediction unit can be set up to predict the delivery time with respect to each of the multitude of routes by adding a required time calculated on the basis of traffic volume and a delay time calculated according to the latency time.

[0024] The device may further include a training unit that is set up to compare the predicted delivery time and an actual delivery time with respect to the finally determined delivery route after completion of the delivery, to update a delivery time prediction model based on the comparison result if the comparison result meets a certain criterion, and to store the comparison result in a database if the comparison result does not meet the certain criterion.

[0025] If traffic volume and latency cannot be calculated due to a sensor defect or error, the delivery time prediction unit can be configured to predict an average delivery time based on current traffic information, including information on overcrowding and people counts during a specific period stored in a database, and current information on the average latency. The delivery route determination unit can be configured to: generate a variety of delivery routes and set priorities based on the predicted average delivery time; and designate an initial delivery route as the first route with the highest priority.and finally, to determine the delivery route as a second route with a next higher priority based on updated information on traffic volume and latency during the delivery process.

[0026] A device and a method for predicting a delivery time and a delivery route based on an estimate of the number of people according to an embodiment are intended for indoor delivery services using robots and can, using an elevator, an entrance and exit gate, a CCTV or the like, calculate a traffic volume and a latency by counting the number of people or the number of items or objects on the robot's movement path and determining the overcrowding of a particular section, predict the delivery time based on the calculated traffic volume and latency and determine a path with the shortest delivery time as the delivery route. BRIEF DESCRIPTION OF THE DRAWINGS Fig. Figure 1 schematically shows a system for predicting a delivery time and a delivery route based on an estimate of the number of people according to one embodiment. Fig. 2 and Fig. Figure 3 shows block diagrams of a device for predicting a delivery time and delivery route based on an estimate of the crowd size according to one embodiment. Fig. 4, Fig. 5 to Fig. Figure 6 shows flowcharts of a method for predicting a delivery time and delivery route based on an estimate of the number of people according to one embodiment. Fig. Figure 7 shows a drawing to illustrate a method for predicting a delivery time and delivery route based on an estimate of the number of people according to one embodiment. Fig. Figure 8 shows a drawing to illustrate a calculating device according to one embodiment. DETAILED DESCRIPTION OF THE EXECUTION FORMS

[0027] One embodiment of the disclosure is described in more detail below with reference to the accompanying drawings, so that a person skilled in the art can easily implement the embodiment. As a person skilled in the art would recognize, the described embodiments can be modified in various ways without deviating from the meaning or scope of the present disclosure. To clarify the present disclosure, parts not related to the description are omitted, and the same elements or equivalents are designated by the same reference numerals throughout the patent specification.

[0028] Furthermore, unless explicitly stated otherwise, the word "show / shows" and variants such as "shows" or "showing" are understood to mean the inclusion of the mentioned elements, but not the exclusion of other elements. Terms encompassing an ordinary number, such as "first" and "second," are used to describe different components, but these terms do not restrict the components. The terms are used only to distinguish one component from other components.

[0029] Furthermore, the terms “unit”, “part” or “section”, “-er” and “module” in the patent specification refer to a unit that processes at least one function or process or operation which can be implemented by hardware, software or a combination of hardware and software.

[0030] The following describes embodiments of the present disclosure with reference to the drawings.

[0031] Fig. Figure 1 schematically shows a system for predicting a delivery time and a delivery route based on an estimate of the number of people according to one embodiment.

[0032] With reference to Fig. 1 A system for predicting a delivery time and route based on an estimate of the crowd may include a device 100 for predicting a delivery time and route based on an estimate of the crowd, a delivery robot 10, an elevator 20, a CCTV 30 and an entrance and exit gate 40.

[0033] The device 100 for predicting a delivery time and route based on an estimate of the crowd, the delivery robot 10, the elevator 20, the CCTV 30 and the entrance and exit gate 40 can be interconnected via a wired / wireless network.

[0034] The device 100 for predicting a delivery time and delivery route based on an estimate of the crowd predicts a delivery time of the delivery robot 10 based on information for estimating the crowd including information on counting people and information on overcrowding and can generate an optimal delivery route based on the predicted delivery time of the robot.

[0035] The delivery robot 10 can be implemented as a robot that performs unmanned delivery within a specific space, including the interior of a building.

[0036] The delivery robot 10 can move within the building according to a stored movement plan.

[0037] The delivery robot 10 can include various types of robot sensors, such as a camera, lidar, or similar.

[0038] A variety of elevators (Elevator 20) may be present along the delivery route within the building. Elevator 20 may include various elevator sensors, including a camera sensor, a thermal imaging sensor, a door opening / closing sensor, and a weight sensor.

[0039] A variety of CCTV 30 cameras may be present along the delivery route within the building. The CCTV 30 camera may include a camera sensor and a thermal imaging sensor.

[0040] Within the building, there may be a variety of entrance and exit gates (40) along the delivery route. These entrance and exit gates (40) can incorporate various sensors to detect and count people or objects passing through them.

[0041] Fig. 2 and Fig. Figure 3 shows block diagrams of a device for predicting a delivery time and delivery route based on an estimate of the crowd size according to one embodiment.

[0042] With reference to Fig. 2 and Fig. 3 The device 100 for predicting a delivery time and delivery route based on an estimate of the crowd may comprise a prediction unit 110, a sensor interface 120, a data analyzer 130, a robot service interface 140 and a training unit 150.

[0043] The device 100 for predicting a delivery time and delivery route based on an estimate of the crowd can be used with the various sensors 50, including a camera sensor, a thermal imaging sensor, a lidar sensor, an elevator sensor, a robot sensor and an entry and exit gate sensor, which are located in the delivery robot 10, the elevator 20, the CCTV 30 and the entry and exit gate 40 of Fig. 1 are attached, and are connected via the sensor interface 120.

[0044] The prediction unit 110 can receive crowd estimate information from the sensor interface 120 and the data analyzer 130 and can predict delivery time based on traffic volume and latency calculated on the basis of the crowd estimate information received.

[0045] Based on the predicted delivery time, the prediction unit 110 can determine the optimal delivery route from a multitude of delivery routes for the robot.

[0046] The prediction unit 110 can estimate the delivery time and delivery route using an artificial intelligence prediction model that has been trained according to repeated prediction of the delivery time and determination of the delivery route.

[0047] In one embodiment, the data analyzer 130 can receive the stored movement path plan from the delivery robot when receiving a delivery request.

[0048] The data analyzer 130 can determine a large number of routes based on the movement plan, where the difference in travel time between the routes to the destination does not exceed a threshold.

[0049] The Data Analyzer 130 can use sensors to gather crowd estimate information for a destination, taking into account the multitude of routes. This crowd estimate can include the level of congestion in a given environment, calculated by a camera and sensor. Alternatively, the crowd estimate can be calculated based on the number of people and the degree of occupancy in a specific area.

[0050] The data analyzer 130 can include a people counter 131, an overcrowding analyzer 132 and the traffic analyzer 133.

[0051] The 131 people counter can collect people counting information for any of its many routes using a camera sensor, a lidar sensor, and a thermal imaging sensor. The people counting information can include the number of people within a specific area, as determined by a camera sensor or similar device.

[0052] The 132 overcrowding analyzer can calculate overcrowding on any of the numerous pathways based on people count information. The more people on the pathway, the greater the potential overcrowding.

[0053] The overcrowding analyzer 132 can detect overcrowding of a specific section, including the elevator section and the entrance and exit gate section, on the pathways based on entry and exit log data obtained from the elevator sensor and the entrance and exit gate sensor.

[0054] Each of the many pathways can contain a multitude of specific sections. The 132 overcrowding analyzer can calculate the overcrowding of the corresponding sections based on real-time recordings of elevator and / or entrance / exit gate entries and exits. That is, the more entries and exits are recorded, the greater the potential overcrowding.

[0055] For example, the overcrowding analyzer 132 can determine that the overcrowding of the corresponding section is greater than that of the other sections if the records of inputs and outputs increase within a preset time.

[0056] For example, if a disabled-accessible speed gate is opened among the entrance and exit gates, or if any entrance and exit gate is open for an extended period of time, the overcrowding analyzer 132 may determine that a large object or a majority of people have entered or left the section or area at the same time.

[0057] The traffic analyzer 133 can calculate the traffic volume based on information about the number of people and the overcrowding of the specific section and store the calculated traffic volume in a database DB.

[0058] Traffic volume can be proportional to the number of people and congestion on the route. Traffic volume is calculated for each of the many routes, and the predicted delivery time for a route with low traffic volume is shorter than the predicted delivery time for a route with high traffic volume.

[0059] Traffic volume data can include assessment information calculated based on people counts and overcrowding data. This assessment information can represent the relative traffic volume between routes.

[0060] The traffic analyzer 133 can determine, based on the number of people and the level of overcrowding, whether it is possible to pass through the elevator or the entrance and exit gate on the route of travel at the time of delivery.

[0061] This means that, based on the crowd estimate information, the traffic analyzer 133 can detect an excessive traffic situation or the like, in which the robot cannot use the elevator or the entrance and exit gate due to a specific event or overcrowding phenomenon along the way.

[0062] The prediction unit 110 can include a path analyzer 111, a delivery time prediction unit 112 and a delivery route determination unit 113.

[0063] Based on crowd estimate information, the route analyzer 111 can calculate the traffic volume and latency to the destination for each of the multitude of routes.

[0064] The path analyzer 111 can calculate the latency based on the time between when the crowd estimation information is acquired and when the traffic volume is calculated using the crowd estimation information.

[0065] Latency can be a factor that delays the predicted delivery time of the robot shipment.

[0066] The delivery time prediction unit 112 can predict the delivery time to the destination based on traffic volume and latency time in relation to each of the multitude of routes.

[0067] With respect to each of the multitude of routes, the delivery time prediction unit 112 can predict the delivery time by adding a required time calculated on the basis of traffic volume and a delay time calculated according to the latency time.

[0068] If the traffic volume and latency cannot be calculated due to a defect or error of the sensor, the delivery time prediction unit 112 can predict an average delivery time for this route based on current traffic information, including information on overcrowding and counting of people during a current specific period of time stored in the database DB, as well as current information on the average latency.

[0069] The delivery route determination unit 113 can determine a route with the shortest predicted delivery time from the multitude of routes as the delivery route.

[0070] If the traffic volume and latency cannot be calculated due to a defect or error of the sensor, the delivery route determination unit 113 can generate a variety of delivery routes based on the predicted average delivery time and can set priorities between the delivery routes.

[0071] The delivery route determination unit 113 can determine an initial delivery route as a first route with the highest priority and can eventually determine the delivery route as a second route with the next highest priority based on real-time updated traffic volume and latency information while the robot is making the delivery.

[0072] This means that the delivery route determination unit 113 can initially exclude routes with long average delivery times from the delivery routes. The delivery route determination unit 113 can initially determine the first route and, if the delivery time of the first route increases by a threshold or more in real time before or during the robot delivery, according to traffic volume and latency, change the delivery route to the second route.

[0073] The prediction unit 110 can provide the user with information about the predicted or expected delivery time and delivery route via a user application or a website of a user terminal device.

[0074] The Robot Service Interface 140 can transmit data to and request data from a control server for a robot delivery service. For example, the Robot Service Interface 140 can request product order details and estimated cooking and preparation times.

[0075] The robot service interface 140 can provide various information regarding orders and deliveries to the prediction unit 110.

[0076] The Robot Service Interface 140 can provide an initial estimated delivery time based on delivery information such as order details and preparation time.

[0077] The robot service interface 140 can supply the training unit 150 with various information to train the prediction model.

[0078] After completion of the delivery, training unit 150 can compare the predicted delivery time with the final determined delivery route and an actual delivery time, update the delivery time prediction model based on the comparison result if the comparison result meets a specific criterion, and store the comparison result in the database DB if the comparison result does not meet the specific criterion.

[0079] The database can store analyzed data regarding overcrowding and traffic volume. The database can also store data on predicted delivery times and specific delivery routes based on order information.

[0080] The database DB can provide the stored data to training unit 150, and training unit 150 can use the provided data to update the prediction model in real time.

[0081] The database DB can be provided separately, without being located on the data analyzer 130.

[0082] Fig. 4, Fig. 5 to Fig. Figure 6 shows flowcharts of a method for predicting delivery time and delivery route based on an estimate of the number of people according to one embodiment.

[0083] The method for predicting delivery time and route based on crowd size estimates Fig. 4, Fig. 5 to Fig. 6 can use device 100 to predict the delivery time and route based on an estimate of the number of people. Fig. 1 will be carried out.

[0084] Fig. Figure 4 shows a flowchart illustrating a process of training the prediction model according to one embodiment.

[0085] In Fig. 4. The device 100 can collect sensor data and infrastructure data to predict the delivery time and route based on the crowd estimate in step S410 when the delivery begins.

[0086] The device 100 for predicting delivery time and delivery route based on crowd estimates can acquire order information, sensor data about the robot's movement path, and infrastructure data via the sensor interface 120 and the robot service interface 140.

[0087] In step S420, the device 100 can analyze crowding and traffic volume along the delivery route to predict delivery time and route based on crowd estimates. Information on people counting, crowding, and traffic volume can be stored in the database in real time.

[0088] In step S430, when the delivery is completed according to the delivery route extracted on the basis of the traffic volume, the device 100 can capture the delivery time and delivery route information based on the crowd estimate.

[0089] In step S440, the device 100 can analyze the delivery time for each delivery route based on the crowd estimate to predict the delivery time and route.

[0090] In step S450, the device 100 can train and update the prediction model based on the analysis data to predict the delivery time and delivery route based on the estimate of the crowd.

[0091] If the validity of the analysis data is taken into account and the data is determined to be valid, the device 100 can update the prediction model for predicting the delivery time and route based on the estimate of the crowd size using the relevant data.

[0092] The device 100 for predicting delivery time and delivery route based on crowd estimates can compare the predicted delivery time and the actual delivery time and can determine the validity of the data based on whether the difference is below a predetermined value.

[0093] Alternatively, the device 100 for predicting the delivery time and delivery route based on the estimate of the crowd can determine it as invalid if the predicted delivery route and / or delivery time has an abnormal value that exceeds a general (predetermined) level or value.

[0094] If the data is invalid, the device 100, which predicts delivery time and route based on the crowd estimate in step S460, stores the data in the database. The data stored in the database can later be used to train the predictive model or to provide information.

[0095] In Fig. 5. The device 100 can set a variety of robot movement paths to the destination for predicting the delivery time and delivery route based on the estimate of the crowd in step S510.

[0096] When the delivery request is received, the device 100 can receive the stored motion path plan from the robot to predict the delivery time and route based on the crowd estimate and can determine the multitude of paths based on the motion path plan where the travel time difference between paths to the destination does not exceed the threshold.

[0097] In step S520, the device 100 can calculate the number of people and the overcrowding of the specified section, including the elevator and the entrance and exit gate, to predict the delivery time and route based on the crowd estimate, and can capture the crowd estimate information in relation to each of the robot's movement paths.

[0098] The device 100 for predicting delivery time and route based on crowd estimates can capture people counting information for each of the multitude of routes via a camera sensor and a thermal imaging sensor.

[0099] The device 100 for predicting delivery time and delivery route based on crowd estimates can detect overcrowding in the specified section, including the elevator section and the entrance and exit gate section, on the routes based on the entry and exit log data obtained from the elevator sensor and the entrance and exit gate sensor.

[0100] In step S530, the device 100 can calculate the traffic volume and latency for each of the multitude of routes based on the crowd estimate information to predict the delivery time and route.

[0101] The device 100 for predicting delivery time and route based on crowd estimates can calculate traffic volume based on information on counting people and overcrowding of the specific section and store the calculated traffic volume in the database.

[0102] The device 100 for predicting delivery time and route based on crowd estimates can calculate the latency based on the time between when the crowd estimate information is acquired and when the traffic volume is calculated using the crowd estimate information.

[0103] In step S540, the device 100 can predict the delivery time and route based on the crowd estimate, predicting the delivery time to the destination based on the traffic volume and latency time with respect to each of the multitude of routes.

[0104] The device 100 for predicting delivery time and delivery route based on crowd estimates can predict the delivery time in respect of each of the multitude of routes by adding the required time calculated on the basis of traffic volume and the delay time calculated according to the latency time.

[0105] In step S550, the device 100 can determine, based on the estimate of the crowd, a route with the shortest delivery time from the multitude of routes as a final delivery route for predicting the delivery time and route.

[0106] After completion of the delivery, the device 100 can compare the predicted delivery time and delivery route based on the estimate of the number of people with the finally determined delivery route and the actual delivery time and, if the comparison result meets the specified criterion, update the model for predicting the delivery time based on the comparison result.

[0107] If the comparison result does not meet the specified criterion, the device 100 can store the comparison result in the database to predict the delivery time and route based on the estimate of the number of people.

[0108] Fig. Figure 6 shows a flowchart illustrating a method for predicting delivery time and route based on an estimate of the crowd size according to one embodiment, when traffic volume and latency cannot be calculated due to a defect or error of the sensor.

[0109] In Fig. 6. The device 100 can determine the multitude of robot movement paths available to the destination for predicting the delivery time and route based on the estimate of the crowd in step S610.

[0110] In step S620, the device 100 can, for predicting the delivery time and route based on the crowd estimate, acquire information on the current traffic volume, including overcrowding, and information on counting people during the current specified time period in relation to the multitude of robot movement paths, as well as information on the current average latency time from the database.

[0111] In step S630, the device 100 can predict the delivery time and route based on the crowd estimate, using information about current traffic volume and information about current average latency to generate the multitude of delivery routes and set priorities.

[0112] In step S640, the device 100 can predict the delivery time and route based on the crowd estimate, determine the initial delivery route as the first route with the highest priority, and can finally determine the delivery route as the second route according to a next higher priority based on traffic and latency information updated during the delivery process.

[0113] Fig. Figure 7 shows a drawing to illustrate a method for predicting the delivery time and delivery route based on the estimate of the number of people according to one embodiment.

[0114] In Fig. 7. The delivery route available from a starting point A to a destination B has a first delivery route PATH1 and a second delivery route PATH2.

[0115] The predicted delivery time (ETA) via the first delivery route, PATH1, can be 12 minutes, taking into account both 10 minutes of traffic and a latency of 2 minutes. In comparison, the predicted delivery time via the second delivery route, PATH2, can be 14 minutes.

[0116] The device 100 for predicting the delivery time and delivery route based on the estimate of the crowd can finally determine the first delivery route PATH1 where the traffic volume determined on the basis of the overcrowding and the delivery time predicted on the basis of the latency are shorter.

[0117] Fig. Figure 8 shows a drawing to illustrate a calculating device according to one embodiment.

[0118] With reference to Fig. 8. A device and a method for predicting the delivery time and delivery route of the robot can be implemented based on the estimation of the crowd according to embodiments using a computer or computing device 900.

[0119] The computing device 900 can comprise at least one processor 910, a memory 930, a user interface input device 940, a user interface output device 950, and a storage device 960, which communicate via a bus 920. The computing device 900 can also comprise a network interface 970, which is electrically connected to a network 90. ​​The network interface 970 can send, transmit, or receive signals with other devices via the network 90.

[0120] The 910 processor can be implemented in various ways, such as a microcontroller unit (MCU), an application processor (AP), a central processing unit (CPU), a graphics processing unit (GPU), a neural processing unit (NPU), and the like, and can be any type of semiconductor device capable of executing instructions stored in the 930 memory or the 960 storage device. The 910 processor can be configured to perform the above-mentioned tasks. Fig. 1, Fig. 2, Fig. 3, Fig. 4, Fig. 5, Fig. 6 to Fig. to implement the 7 described functions and procedures.

[0121] The memory 930 and the memory device 960 can comprise various types of volatile or non-volatile storage media. For example, the memory can comprise a read-only memory (ROM) 931 and a random access memory (RAM) 932. In this embodiment, the memory 930 can be located inside or outside the processor 910, and the memory 930 can be connected to the processor 910 by various known means.

[0122] In some embodiments, at least some configurations or arrangements or functions of a device and a method for predicting a robot's delivery time and delivery route based on the estimate of the crowd according to an embodiment can be implemented as a program or software that is executable by the computing device 900, and the program or software can be stored in a computer-readable medium.

[0123] In some embodiments, at least some configurations or arrangements or functions of a device and a method for predicting a robot's delivery time and a robot's delivery route based on the estimate of the crowd according to an embodiment can be implemented using hardware or circuits of the computing device 900, or they can also be implemented as separate hardware or circuits that may be electrically connected to the computing device 900.

[0124] Although this disclosure has been described in connection with the embodiments currently considered practical, it is to be understood that the disclosure is not limited to the disclosed embodiments, but on the contrary is intended to cover various modifications and equivalent arrangements which are included in the sense and scope of the attached claims. <Beschreibung der Bezugszeichen> 100 Device for predicting delivery time and route based on an estimate of the number of people 110 forecast unit 120 sensor interface 130 Data analyzer 140 Robot service interface 150 training units

Claims

(Originally) A method for predicting delivery time and route based on crowd estimates, comprising: defining a plurality of routes to reach a destination; collecting crowd estimate information on the plurality of routes by sensors on the plurality of routes; calculating traffic volume and latency for each of the plurality of routes based on the crowd estimate information; predicting the delivery time to the destination based on the traffic volume and latency with respect to each of the plurality of routes; and determining the delivery route as one with the shortest delivery time among the plurality of routes. (Currently amended) Method according to claim 1, wherein determining the plurality of paths to reach the destination comprises receiving a stored motion path plan from a robot when a delivery request is received, and determining the plurality of paths based on the motion path plan, wherein a path time difference to the destination between paths does not exceed a threshold. (Currently amended) Method according to claim 1, wherein the sensors comprise a camera sensor comprising a CCTV, a lidar sensor, a thermal imaging sensor, an elevator sensor, a robot sensor and an entry and exit gate sensor. (Originally) Method according to claim 3, wherein the acquisition of information for estimating the crowd size of the plurality of paths comprises: acquiring information for counting people for each of the plurality of paths by the camera sensor and the thermal imaging sensor. (Originally) The method of claim 4, wherein the acquisition of information for estimating the crowd size of the plurality of paths comprises: detecting an overcrowding of a particular section comprising an elevator section and an entry and exit gate section on the paths, based on entry and exit log data obtained from the elevator sensor and the entry and exit gate sensor. (Original) Method according to claim 5, wherein the calculation of the traffic volume and latency for each of the plurality of paths based on the crowd estimation information comprises: calculating the traffic volume based on the information on counting people and the overcrowding of the specified section and storing the calculated traffic volume in a database. (Original) Method according to claim 6, wherein, if the traffic volume and latency cannot be calculated due to a defect or error of the sensor, the prediction of the delivery time to the destination based on the traffic volume and latency with respect to each of the plurality of routes further comprises a prediction of an average delivery time based on current traffic information, which includes information on overcrowding and counting of persons during a current specific period of time stored in the database, and current information on the average latency, and the determination of the delivery route as the route with the shortest delivery time among the plurality of routes: generating a plurality of delivery routes and setting priorities based on the average delivery time;and determining an initial delivery route as a first route with the highest priority and finally determining the delivery route as a second route with the next highest priority based on traffic and latency information updated during the delivery process. (Original) Method according to claim 1, wherein the calculation of the traffic volume and latency for each of the plurality of paths based on the crowd estimation information comprises: calculating the latency based on a time between a time at which the crowd estimation information is acquired and a time of calculating the traffic volume using the crowd estimation information. (Original) Method according to claim 1, wherein the prediction of the delivery time to the destination based on the traffic volume and the latency time with respect to each of the plurality of routes comprises: Predicting the delivery time with respect to each of the plurality of routes by adding a required time calculated on the basis of the traffic volume and a delay time calculated according to the latency time. (Currently amended) Method according to claim 1, further comprising: comparing the predicted delivery time and an actual delivery time with respect to the specified delivery route after completion of a delivery, and updating a delivery time prediction model based on a comparison result if the comparison result meets a specified criterion; and storing the comparison result in a database if the comparison result does not meet the specified criterion. (Originally) Device for predicting a delivery time and delivery route based on a crowd estimate, comprising: a data analyzer configured to acquire crowd estimate information to a destination with respect to a plurality of routes by means of a sensor; a route analyzer configured to calculate, based on the crowd estimate information, the traffic volume and latency to the destination with respect to each of the plurality of routes; a delivery time prediction unit configured to predict the delivery time to the destination based on the traffic volume and latency with respect to each of the plurality of routes; and a delivery route determination unit configured to determine the delivery route as one with the shortest delivery time among the plurality of routes. (Original) Device according to claim 11, wherein, when a delivery request is received, the data analyzer is configured to receive a stored motion path plan from a robot and to determine the plurality of paths based on the motion path plan where a path time difference to the destination between paths does not exceed a threshold. (Originally) Device according to claim 11, wherein the sensor comprises a CCTV camera sensor, a lidar sensor, a thermal imaging sensor, an elevator sensor, a robot sensor and an entry and exit gate sensor. (Originally) Device according to claim 13, wherein the data analyzer is configured to acquire information for counting persons for each of the plurality of paths through the camera sensor and the thermal imaging sensor. (Original) Device according to claim 14, wherein the data analyzer is configured to detect an overcrowding of a particular section comprising an elevator section and an entry and exit gate section on the pathways, based on entry and exit log data obtained from the elevator sensor and the entry and exit gate sensor. (Originally) Device according to claim 15, wherein the path analyzer is configured to calculate the traffic volume based on information on the counting of persons and the overcrowding of the specified section and to store the calculated traffic volume in a database. (Original) Device according to claim 11, wherein the path analyzer is configured to calculate the latency based on a time between a time at which the crowd estimation information is acquired and a time at which the traffic volume is calculated using the crowd estimation information. (Originally) Device according to claim 11, wherein the delivery time prediction unit is configured to predict the delivery time with respect to each of the plurality of routes by adding a required time calculated on the basis of traffic volume and a delay time calculated according to the latency time. (Originally) Device according to claim 11, further comprising a training unit configured to compare the predicted delivery time and an actual delivery time with respect to the finally determined delivery route after completion of delivery, to update a delivery time prediction model based on the comparison result if the comparison result meets a certain criterion, and to store the comparison result in a database if the comparison result does not meet the certain criterion. (Originally) Device according to claim 11, wherein, if the traffic volume and latency cannot be calculated due to a defect or error of the sensor, the delivery time prediction unit is configured to predict an average delivery time based on current traffic information, which includes information on overcrowding and counting of persons during a current specific period of time stored in a database, and information on the current average latency, and the delivery route determination unit is configured to: generate a plurality of delivery routes and establish priorities based on the predicted average delivery time;to determine an initial delivery route as the first route with the highest priority, and finally to determine the delivery route as the second route with the next highest priority based on the traffic and latency information updated during the delivery process.

Images