Elevator dispatch arbitration method and system

Abstract

The present application relates to elevator dispatching technical field, disclose a kind of elevator dispatching arbitration method, comprising the following steps: step 1: receiving multi-source call request, corresponding identity fingerprint is assigned to each call request and monitors the waiting time of each call request;Step 2: obtain the current load of target elevator and car remaining space, whether the current load of target elevator and car remaining space meets corresponding call request is judged;When the current load of target elevator or car remaining space does not meet corresponding call request, then cancel the call request;When the current load of target elevator and car remaining space both meet corresponding call request, then carry out step 3;Step 3: using dispatching score model to calculate call request, obtain the score of call request;Step 4: using shadow dispatcher to send shadow instruction to target elevator, shadow instruction triggers the corresponding stop layer action of target elevator.Simultaneously, a kind of elevator dispatching arbitration system is also disclosed.

Description

Technical Field

[0001] This application belongs to the field of elevator dispatching technology, and more specifically, relates to an elevator dispatching arbitration method and system. Background Technology

[0002] With the development of smart buildings, elevator call requests no longer come only from traditional physical buttons, but have expanded to various terminals such as indoor video intercom extensions, facial recognition door stations, access control readers, and autonomous delivery robots.

[0003] In some elevator scheduling solutions, when people and robots share an elevator, the elevator is controlled solely based on the order in which calls are made. This can lead to situations where the remaining space in the elevator is insufficient for the robot to enter, causing the robot to block the doorway and preventing the elevator from operating normally, thus reducing the riding experience. In other elevator scheduling solutions, the remaining space in the elevator is assessed to determine if the robot can enter, and priority is set to control whether the elevator stops at a floor. However, this method may result in the robot's size repeatedly failing to meet the remaining space in the elevator, or the robot's low priority causing it to wait in place and be unable to complete its work.

[0004] Therefore, the technical problem solved by this application is: how to improve the scheduling efficiency of elevators to meet the elevator usage needs of various terminals. Summary of the Invention

[0005] The main purpose of this application is to provide an elevator scheduling arbitration method. By assigning a corresponding identity fingerprint to each elevator call request, and based on the identity fingerprint, determining whether the remaining space of the elevator can meet the elevator call request, and then scoring all elevator call requests based on waiting time and initial priority, the elevator stops at the floor according to the score. In this way, the elevator needs of humans and robots can be balanced while minimizing the waiting time of various terminals.

[0006] In addition, an elevator dispatch arbitration system is also provided.

[0007] To achieve the above objectives, the technical solution adopted in this application is as follows:

[0008] An elevator dispatch arbitration method includes the following steps:

[0009] Step 1: Receive multi-source elevator call requests, assign a corresponding identity fingerprint to each elevator call request and monitor the waiting time of each elevator call request, and then include the identity fingerprint in the form of identity attributes, initial priority attributes, location attributes and physical attributes;

[0010] Step 2: Obtain the current load and remaining car space of the target elevator. Based on the identity attribute and physical attribute, determine whether the current load and remaining car space of the target elevator meet the corresponding call request. If the current load or remaining car space of the target elevator does not meet the corresponding call request, cancel the call request. If the current load and remaining car space of the target elevator both meet the corresponding call request, proceed to Step 3.

[0011] Step 3: Based on the initial priority attribute, waiting time and location attribute, the scheduling scoring model is used to calculate the elevator call request and obtain the elevator call request score;

[0012] Step 4: Based on the scoring of the elevator call request, a shadow dispatcher is used to send a shadow command to the target elevator, which triggers the corresponding stop action of the target elevator.

[0013] Preferably, in step 1, the identity attributes include four types of identity requests: the first type of identity request includes fire-fighting requests and emergency medical requests; the second type of identity request includes facial recognition requests and access control card swiping requests; the third type of identity request includes logistics robot requests and delivery robot requests; and the fourth type of identity request includes indoor unit requests and APP remote reservation requests.

[0014] The initial priority attribute includes the initial score of the first type of identity request, the initial score of the second type of identity request, the initial score of the third type of identity request, and the initial score of the fourth type of identity request; the initial score of the first type of identity request > the initial score of the second type of identity request > the initial score of the third type of identity request > the initial score of the fourth type of identity request.

[0015] The location attribute is the floor where the elevator call request is located;

[0016] The physical properties include weight and geometric dimensions.

[0017] Preferably, step 2 includes the following sub-steps:

[0018] Step A1: Obtain the current load and remaining space in the car of the target elevator;

[0019] Step A2: Based on the identity attribute, determine whether the elevator call request belongs to the third type of identity request. If the elevator call request does not belong to the third type of identity request, proceed to step 3; if the elevator call request belongs to the third type of identity request, proceed to step A3.

[0020] Step A3: Set the load threshold and, based on the physical size attribute, determine whether the current load of the target elevator is less than the load threshold and whether the remaining space of the target elevator car is greater than the physical size attribute; if the current load of the target elevator is not less than the load threshold, or the remaining space of the target elevator car is not greater than the physical size attribute, then the stop action is not triggered; if the current load of the target elevator is less than the load threshold and the remaining space of the target elevator car is greater than the physical size attribute, then proceed to step 3.

[0021] Preferably, step 3 includes the following sub-steps:

[0022] Step B1: Obtain the initial score corresponding to the elevator call request based on the initial priority attribute of the elevator call request;

[0023] Step B2: Based on the waiting time of the elevator call request, a step gain mechanism is used to judge the waiting time and obtain the time gain score of the elevator call request;

[0024] Step B3: Based on the target elevator status and the location attributes of the floor and the call request, obtain the compatibility coefficient;

[0025] Step B4: Based on the energy consumption of the distance the target elevator moves to the call request and the energy consumption of the switching time, obtain the marginal stop cost;

[0026] Step B5: Input the initial score, time gain score, compatibility coefficient, and marginal stop cost into the scheduling score model to obtain the score for the elevator call request.

[0027] Preferably, in step B2, the step gain mechanism is specifically as follows:

[0028] When the waiting time for an elevator call request is less than the first time threshold, the time gain score is 0.

[0029] When the first time threshold is less than or equal to the waiting time for the elevator call request and less than the second time threshold, the time gain score is the first set value.

[0030] When the second time threshold is less than or equal to the waiting time for the elevator call request and less than the third time threshold, the time gain score is the second set value.

[0031] When the third time threshold is less than the waiting time for the elevator call request, the time gain score is the third set value.

[0032] Preferably, the formula for the scheduling scoring model is:

[0033]

[0034] Where S is the rating of the elevator call request. The initial score corresponding to the elevator call request. For compatibility coefficient, Value for marginal stops Forward distance reward.

[0035] Preferably, the method further includes step 5: when the identity attribute of the elevator call request is a fourth type of identity request, the shadow dispatcher sends a shadow instruction to the target elevator and starts the reservation timer. When the reservation timer reaches the preset buffer time and no passenger making the elevator call request is detected, the shadow dispatcher sends a reverse intervention instruction to cancel the corresponding shadow instruction.

[0036] In addition, an elevator dispatching arbitration system is provided to implement the above-mentioned elevator dispatching arbitration method, comprising the following units:

[0037] Elevator call request receiving unit: used to receive multi-source elevator call requests, assign a corresponding identity fingerprint to each elevator call request and monitor the waiting time of each elevator call request, and then include the identity fingerprint including identity attributes, initial priority attributes, location attributes and physical attributes;

[0038] Elevator call request judgment unit: used to obtain the current load and remaining car space of the target elevator, and determine whether the current load and remaining car space of the target elevator meet the corresponding elevator call request based on the identity attribute and physical attribute; if the current load or remaining car space of the target elevator does not meet the corresponding elevator call request, the elevator call request is cancelled; if the current load and remaining car space of the target elevator both meet the corresponding elevator call request, then proceed to step 3.

[0039] Elevator call request scoring unit: It is used to calculate the elevator call request score based on the initial priority attribute, waiting time and location attribute using the scheduling scoring model;

[0040] Elevator stop control unit: Used to score based on elevator call requests, and uses a shadow dispatcher to send shadow commands to the target elevator, which triggers the corresponding stop action of the target elevator.

[0041] One of the above-mentioned technical solutions in this application has at least one of the following advantages or beneficial effects:

[0042] The elevator scheduling arbitration method of this application assigns a corresponding identity fingerprint to each elevator call request. Based on the identity fingerprint, it first determines whether the remaining space in the elevator can meet the elevator call request. Then, it scores all elevator call requests according to the waiting time and initial priority. The elevator stops at the designated floors based on the scores. In this way, under the control of the initial priority, users (humans) can take the elevator as soon as possible with a higher initial priority, improving the user (human) elevator experience. However, under the premise of a lower initial priority and size limitations, the elevator call requests of robots may be suspended indefinitely. Therefore, by setting a waiting time and using a scheduling scoring model to score elevator call requests, it is ensured that the elevator can prioritize meeting the elevator call requests of robots after a long waiting time. This achieves a balance between the elevator needs of humans and robots while minimizing the waiting time of various terminals.

[0043] Secondly, the elevator scheduling model calculates based on four dimensions: initial score, time gain score, compatibility coefficient, and marginal stop cost. In this way, each elevator call request is not only judged based on the initial priority, but also comprehensively scored by combining the initial priority, waiting time, forward stopping, and stopping energy consumption. This allows for a better comprehensive analysis of each elevator call request and avoids situations such as forward stopping being ignored due to low initial priority, robot elevator call requests being permanently blocked by high-frequency traffic, and elevators rigidly moving back and forth between high-priority floors, resulting in excessively long waiting times for low-priority users on other floors.

[0044] Secondly, for the fourth type of identity request (reservation for elevator ride), a reservation timer has been added. When the preset buffer time is reached and no passenger with the elevator call request is detected, the shadow dispatcher sends a reverse intervention command to cancel the corresponding shadow command, thereby avoiding the elevator running empty and improving the elevator dispatching efficiency. Attached Figure Description

[0045] The present application will be further described below with reference to the accompanying drawings and embodiments;

[0046] Figure 1 Here is a flowchart of an elevator scheduling arbitration method according to Example 1;

[0047] Figure 2 This is a block diagram of an elevator dispatching arbitration system according to Example 2. Detailed Implementation

[0048] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain this application, and should not be construed as limiting this application.

[0049] The following disclosure provides many different implementation methods or examples for different schemes of implementing this application.

[0050] Example 1

[0051] refer to Figure 1 An elevator dispatch arbitration method includes the following steps:

[0052] Step 1: Receive multi-source elevator call requests, assign a corresponding identity fingerprint to each elevator call request and monitor the waiting time of each elevator call request, and then include the identity fingerprint in the form of identity attributes, initial priority attributes, location attributes and physical attributes;

[0053] In this embodiment, the multi-source elevator call request includes original requests using SIP, MQTT, Wiegand, and HTTP protocols. The first type of identity request mainly uses SIP protocol (for emergency voice and command linkage) or dry node / physical level signal (to ensure the highest physical priority). The second type of identity request mainly uses Wiegand protocol (direct connection to traditional access control) or HTTP / HTTPS protocol (API push after face recognition). The third type of identity request mainly uses MQTT protocol (IoT standard, supporting low-latency bidirectional asynchronous communication). The fourth type of identity request mainly uses HTTP protocol (RESTful API requests commonly used on mobile devices).

[0054] The gateway's built-in protocol adapter layer unifies multi-source call requests into task metadata. Specifically, it defines a standard schema in JSON format and uses a message broker (such as Mosquitto or RabbitMQ) to parse heterogeneous protocols into a unified structure containing Identity_ID, Target_Floor, and Timestamp—that is, the task metadata.

[0055] A dynamic timer is used at the software level. When a call request is written to the cache queue, the gateway system kernel records a Unix timestamp and calculates the millisecond-level offset in real time using "current system time - initial request time" to monitor the waiting time of each call request.

[0056] The identity attributes include four types of identity requests: the first type includes fire-fighting requests and emergency medical requests; the second type includes facial recognition requests and access control card swipe requests; the third type includes logistics robot requests and delivery robot requests; and the fourth type includes indoor unit requests and APP remote reservation requests.

[0057] The initial priority attribute includes the initial score of the first type of identity request, the initial score of the second type of identity request, the initial score of the third type of identity request, and the initial score of the fourth type of identity request; the initial score of the first type of identity request > the initial score of the second type of identity request > the initial score of the third type of identity request > the initial score of the fourth type of identity request.

[0058] The initial priority attribute is divided according to different caller identities, specifically:

[0059] The initial score for the first type of identity request (fire and emergency calls) is the highest level. Subsequent scheduling optimization models cannot assign a score higher than this initial score to any other elevator call request to ensure that the first type of identity request has immediate response rights. Therefore, in this embodiment, the initial score for the first type of identity request is specifically 1000 points, set based on the highest priority to ensure immediate response in emergency situations. Even if other requests achieve the maximum time gain (80 points) and the maximum compatibility coefficient (1.5), the highest possible score is only about 195 points, far lower than the lowest score of 400 points for the first type under the reverse penalty (coefficient 0.4).

[0060] The initial score for the second type of identity requests (facial recognition requests and access control card swipe requests) is the second highest level, representing "the person is already at the elevator entrance and a quick response is needed." To improve the user's (human's) elevator experience and reduce waiting time, in this embodiment, the initial score for the second type of identity requests is specifically 100 points, set based on the second priority, representing the immediate need that "the person has arrived at the elevator entrance." These requests require a rapid response to improve user experience and avoid users waiting for extended periods in the elevator lobby.

[0061] The initial score for the third type of identity request (logistics robot requests and delivery robot requests) is the third highest level. Since robots lack human emotions and primarily transport goods, their response speed can be slightly slower. In this embodiment, the initial score for the third type of identity request is specifically 70 points, set based on the third priority, allowing the robot to wait appropriately. Setting the initial score to 70 points (instead of 50 points) is to enable the robot to surpass human requests through a distance reward mechanism in forward-moving scenarios, achieving optimized interception along the route.

[0062] The initial score for the fourth type of identity request (indoor unit request and APP remote reservation request) is the lowest. Since the fourth type of identity request is a reservation call request, the user will arrive close to the reservation time. Therefore, the initial score can be lower, and the first three types of identity requests are processed first. In this embodiment, the initial score for the fourth type of identity request is 30 points. The basis for setting it is: the lowest priority. Reservation requests usually have advance time. It takes time for the user to arrive at the elevator hall after initiating the reservation. Therefore, it can be flexibly scheduled.

[0063] The location attribute is the floor where the elevator call request is located;

[0064] The physical properties include weight and geometric dimensions.

[0065] In this embodiment, since the second type of identity request needs to be compared with the elevator's load capacity and remaining space, the physical attributes are mainly for the second type of identity request. Other identity requests may not have physical attributes set or may have empty physical attributes.

[0066] It should be noted that:

[0067] SIP Protocol: SIP (Session Initiation Protocol) is an application-layer signaling control protocol used to create, modify, and release sessions with one or more participants. These sessions can be Internet multimedia conferencing, IP telephony, or multimedia distribution. Participants in a session can communicate via multicast, unicast, or a combination of both.

[0068] MQTT Protocol: MQTT is a lightweight messaging protocol based on the publish / subscribe pattern.

[0069] Wiegand Protocol: The Wiegand protocol is a communication protocol developed by Motorola, primarily used for data transmission between card readers and cards in access control systems. This protocol defines a binary transmission method based on two data lines (DATA0 and DATA1), using negative pulse signals to represent 0 and 1 respectively, with a pulse width of 100 microseconds and an interval period of 1600 microseconds.

[0070] HTTP Protocol: HTTP (HyperText Transfer Protocol) is an application layer protocol used for distributed, collaborative, and hypermedia information systems, forming the foundation of data communication on the World Wide Web (WWW).

[0071] Indoor unit: refers to a video intercom terminal device installed inside a residence or office, also known as an "indoor extension" or "indoor video intercom".

[0072] Step 2: Obtain the current load and remaining car space of the target elevator. Based on the identity attribute and physical attribute, determine whether the current load and remaining car space of the target elevator meet the corresponding call request. If the current load or remaining car space of the target elevator does not meet the corresponding call request, cancel the call request. If the current load and remaining car space of the target elevator both meet the corresponding call request, proceed to Step 3.

[0073] In this embodiment, since the robot cannot determine whether it can enter the elevator, it will block the elevator entrance and affect elevator operation if it cannot enter. Therefore, before the scheduling and scoring model scores the robot (second type of identity request), it is necessary to determine whether the robot's physical attributes meet the current load capacity of the elevator and the remaining space in the car.

[0074] Preferably, step 2 includes the following sub-steps:

[0075] Step A1: Obtain the current load and remaining space of the target elevator. Specifically: Use the load-weighing sensors that come with the elevator, usually installed at the bottom of the car or the end of the steel cable, to read the load data via RS485 protocol or CAN bus to obtain the current load of the target elevator; Use a 3D depth camera (such as a TOF sensor) installed at the top of the car to collect the depth data inside the car in real time. After coordinate transformation to generate a 3D point cloud, use background modeling and dynamic segmentation algorithms to extract the occupied area, and then calculate the real-time available space volume inside the car to obtain the remaining space of the car.

[0076] Step A2: Based on the identity attribute, determine whether the elevator call request belongs to the third type of identity request. If the elevator call request does not belong to the third type of identity request, proceed to step 3; if the elevator call request belongs to the third type of identity request, proceed to step A3.

[0077] In this embodiment, the third type of identity request is a robot, while other identity requests are human. When a human encounters a fully loaded elevator or insufficient space, they can choose not to enter the elevator. However, a robot cannot make this judgment on its own. Therefore, before using the scheduling scoring model to score the third type of identity request, it is necessary to determine whether the elevator load and remaining space meet the needs of the third type of identity request. This is to avoid a situation where the third type of identity request scores highly but the robot blocks the doorway due to insufficient space or load.

[0078] Step A3: Set the load threshold and, based on the physical size attribute, determine whether the current load of the target elevator is less than the load threshold and whether the remaining space of the target elevator car is greater than the physical size attribute; if the current load of the target elevator is not less than the load threshold, or the remaining space of the target elevator car is not greater than the physical size attribute, then the stop action is not triggered; if the current load of the target elevator is less than the load threshold and the remaining space of the target elevator car is greater than the physical size attribute, then proceed to step 3.

[0079] In this embodiment, the load threshold is 60% of the maximum load. When the current load of the target elevator is less than the load threshold and the remaining space of the target elevator car is greater than the physical size attribute, step 3 is performed to score the elevator call request. When the current load of the target elevator is not less than the load threshold, or the remaining space of the target elevator car is not greater than the physical size attribute, the stop action is not triggered (the elevator call request in the target elevator is canceled), and the robot is guided to other idle elevators or freight elevators.

[0080] Step 3: Based on the initial priority attribute, waiting time and location attribute, the scheduling scoring model is used to calculate the elevator call request and obtain the elevator call request score;

[0081] Preferably, step 3 includes the following sub-steps:

[0082] Step B1: Obtain the initial score corresponding to the elevator call request based on the initial priority attribute of the elevator call request;

[0083] Step B2: Based on the waiting time of the elevator call request, a step gain mechanism is used to judge the waiting time and obtain the time gain score of the elevator call request;

[0084] The step gain mechanism is as follows:

[0085] When the waiting time for an elevator call request is less than the first time threshold, the time gain score is 0, which represents the normal waiting time.

[0086] When the first time threshold is less than or equal to the waiting time of the elevator call request and less than the second time threshold, the time gain score is set to a first preset value. In this embodiment, the first preset value is 20 points, indicating a longer waiting time and improved competitiveness within the same category. For example, the third category (70 points) becomes 90 points after adding 20 points, which is still lower than the second category (100 points), but higher than other requests within the same category that did not wait.

[0087] When the second time threshold is less than or equal to the waiting time for the elevator call request and less than the third time threshold, the time gain score is the second set value. In this embodiment, the second set value is 50 points, which represents a long waiting time, thus enabling queue-jumping response. For example, the third category (70 points) plus 50 points equals 120 points, exceeding the initial score of the second category (100 points), allowing it to compete with the second category.

[0088] When the third time threshold is less than the waiting time for the elevator call request, the time gain score is the third set value. In this embodiment, the third set value is 80 points. If the waiting time is too long, a fast response is needed to avoid indefinite suspension. For example, the third category (70 points) plus 80 points equals 150 points, which significantly exceeds the second category (100 points), thus obtaining higher priority.

[0089] It should be noted that,

[0090] In this embodiment, the first time threshold is 60 seconds, the second time threshold is 90 seconds, and the third time threshold is 120 seconds.

[0091] Design basis:

[0092] 60-second threshold: Based on elevator industry user experience standards, a response time within 60 seconds is considered an excellent service level;

[0093] 90-second threshold: This corresponds to the upper limit of a "good" service level. If this time is exceeded, the service needs to be prioritized.

[0094] 120-second threshold: This corresponds to the upper limit of the "acceptable" service level; responses must be prioritized if this time is exceeded.

[0095] Gain increments of 20 / 50 / 80 ensure that long-waiting requests can be gradually elevated to higher priority.

[0096] The time gain mechanism ensures that low-priority requests (such as bots) are not permanently blocked due to frequent high-priority requests. After waiting for 120 seconds, the third type of request (70+80=150 points) can even surpass the second type of request (100 points), achieving "starvation prevention" protection.

[0097] Step B3: Based on the target elevator status and the location attributes of the floor and the call request, obtain the compatibility coefficient;

[0098] In this embodiment, the target elevator status and floor are obtained through the elevator control system interface, specifically: Method 1 (CAN bus communication): Connecting to the CAN bus (Controller Area Network) of the elevator group control system, using CANopen or a proprietary CAN protocol, reading position encoder data or floor sensor signals, reading status registers (up / down / stationary / open / closed / fault, etc.), and reading direction flag bits to obtain the target elevator's current floor, operating status, and operating direction; Method 2 (RS485 serial communication): Connecting to the elevator control cabinet via an RS485 bus, using Modbus RTU or elevator manufacturer's proprietary protocol (such as Otis, Mitsubishi, Hitachi, etc.), periodically polling the elevator controller's status register to obtain relevant data.

[0099] Next, based on the compatibility coefficient table, the target elevator status and its current floor are judged in relation to the location attributes of the call request to obtain the compatibility coefficient; the compatibility coefficient table is shown in Table 1:

[0100] Table 1 Compatibility Coefficient Table

[0101] Forward Rewards 1.5 The floor to be called is located ahead of the elevator's direction of travel, and the direction is the same. Increasing the coefficient improves the score, inducing elevators to stop along the route, improving overall operating efficiency, and reducing the number of reversals. neutral response 1.0 The elevator is stationary, or you request to be on the same floor as the elevator. No gain or penalty; calculation is performed normally based on the initial score and time gain. Reverse punishment 0.4 Calling an elevator requires the elevator to change its current direction of travel. Reducing the coefficient lowers the score and prevents frequent reversals from causing a surge in elevator energy consumption and a decrease in operating efficiency.

[0102] Example explanation:

[0103] The elevator is descending on the 8th floor, and there is a request to descend from the 5th floor → Forward bonus (1.5 coefficient);

[0104] The elevator is going down on the 8th floor, and there is an upward request on the 10th floor → Reverse penalty (coefficient 0.4).

[0105] The elevator is stationary on the 5th floor. There is a request on the 5th floor → Neutral response (coefficient 1.0).

[0106] Step B4: Based on the energy consumption of the distance the target elevator moves to the call request and the energy consumption of the switching time, obtain the marginal stop cost;

[0107] In this embodiment, the energy consumption for moving the target elevator to the distance requested by the elevator call is calculated as follows:

[0108] The formula for calculating distance energy consumption is:

[0109] Distance energy consumption = |target floor - current floor| × energy consumption coefficient per unit floor = |target floor - current floor| × 2;

[0110] Parameter description:

[0111] Energy consumption coefficient per floor: 2 points / floor. Basis for setting: Each floor movement consumes a certain amount of electricity and time, which is set at a cost of 2 points. The greater the distance, the higher the cost.

[0112] Linear relationship: Linear calculation is used, which is simple, intuitive, and consistent with actual energy consumption characteristics;

[0113] Calculation example:

[0114] The elevator is on the 8th floor, and the request is made to be on the 5th floor: Distance energy consumption = |5-8| × 2 = 6 minutes;

[0115] The elevator is on the 8th floor, and the request is made on the 3rd floor: Distance energy consumption = |3-8| × 2 = 10 minutes;

[0116] The elevator is on the 8th floor. The request is to use the elevator on the 8th floor: Distance energy consumption = |8-8| × 2 = 0 points;

[0117] Switching time energy consumption is shown in Table 2: Table 2 Switching Time Energy Consumption

[0118] Human passengers (first, second, and fourth application requests) 5 points Approximately 5-8 seconds Humans move up and down elevators quickly, the standard door opening and closing cycle is short, and the cost can be set relatively low. Robot (Third Type Identity Request) 10 points Approximately 15-25 seconds Robots take longer to enter and exit elevators, including processes such as positioning, turning, entering, and adjusting posture. A higher cost is set to reflect this additional time expense.

[0119] Design considerations:

[0120] Authenticity: Reflects the actual difference in door opening and closing time; the robot does indeed require a longer time.

[0121] Fairness: By differentiating the value of the robots, we can prevent frequent robot downtime from affecting overall efficiency;

[0122] Balance: The cost difference (10 vs 5) is moderate and will not completely suppress bot requests;

[0123] Marginal stop cost = energy consumption of the distance the target elevator moves to the call request + energy consumption of the switching time;

[0124] Calculation example:

[0125] The elevator is on the 8th floor. The robot on the 5th floor is requesting: = |5-8| × 2 + 10 = 6 + 10 = 16 points;

[0126] The elevator is on the 8th floor. A person on the 3rd floor requests: = |3-8| × 2 + 5 = 10 + 5 = 15 points;

[0127] The elevator is on the 8th floor. A person on the 8th floor requests: = |8-8| × 2 + 5 = 0 + 5 = 5 points;

[0128] The forward distance bonus is calculated as follows: 50 points are awarded for forward distance when all of the following conditions are met:

[0129] (1) The request is located in front of the elevator in the direction of travel (forward); (2) Among all forward requests, this request is closest to the elevator.

[0130] Design purpose:

[0131] Improve efficiency: Encourage elevators to prioritize requests that are convenient and close by, thereby reducing the total travel distance;

[0132] Robot optimization: Enables robots to exceed human requests in forward-facing scenarios, achieving the goal of picking up robots along the way;

[0133] Energy optimization: Reduce the number of elevator reversals to lower energy consumption.

[0134] Step B5: Input the initial score, time gain score, compatibility coefficient, and marginal stop cost into the scheduling score model to obtain the score for the elevator call request.

[0135] The formula for the scheduling scoring model is as follows:

[0136] ;

[0137] Where S is the rating of the elevator call request. The initial score corresponding to the elevator call request. For compatibility coefficient, Value for marginal stops Forward distance reward.

[0138] By comprehensively calculating the scores of each elevator call request through the scheduling scoring model, the situation of each elevator call request can be analyzed more accurately, thereby enabling more reasonable elevator calls.

[0139] Take one scenario as an example: the elevator is descending on the 8th floor, a robot on the 5th floor requests to descend, and a human passenger on the 3rd floor requests to descend;

[0140] Initial conditions:

[0141] (1) Current elevator position: 8th floor; Elevator operating status: Downward; Elevator load: 0kg (empty); Remaining space in the car: Sufficient, meeting the requirements for robot elevator use;

[0142] (2) Robot request: 5th floor, downhill direction, waiting time < 60 seconds;

[0143] (3) Human request: 3rd floor, downhill direction, waiting time < 60 seconds;

[0144] Scheduling objective: Calculate the scores of the two requests and determine which request the elevator should respond to first;

[0145] Robot requests rating calculation (5th floor):

[0146] Determine the initial priority score ( ): Robots fall under the third category of identity requests. = 70 points;

[0147] Calculate the time gain score ( ):

[0148] If the waiting time is less than 60 seconds (the first time threshold), according to the stepped gain mechanism, the time gain score is 0 when the waiting time is less than 60 seconds, so γ_time = 0 points.

[0149] Determine the compatibility factor (f(Direction)):

[0150] The elevator is descending on the 8th floor, and the robot requests to descend on the 5th floor. The 5th floor is below the 8th floor and the direction is the same (both are descending). According to Table 1, this is determined to be a "forward reward" scenario, so f(Direction) = 1.5.

[0151] Calculate the marginal cost of stopping ( Distance energy consumption = |5 - 8| × 2 = 3 × 2 = 6 minutes. According to Table 2, door opening and closing time energy consumption = 10 minutes (robot). = 6 + 10 = 16 points;

[0152] Calculate the forward distance reward (Bonus_distance):

[0153] If the robot's request is located in front of the elevator's direction of travel (forward), and among all forward requests (robot on the 5th floor, human on the 3rd floor), the 5th floor is closer to the 8th floor (3rd floor < 5th floor), thus meeting the reward condition, then Bonus_distance = 50 points;

[0154] Calculate the final score (S)

[0155] = (70 + 0) × 1.5 - 16 + 50 = 139 points;

[0156] The robot requested a final score of 139.

[0157] Human passenger request rating calculation (3rd floor)

[0158] Determine the initial priority score ( ):

[0159] Human passengers (using facial recognition or access control card swipes) fall under the second category of identity requests. = 100 points;

[0160] Calculate the time gain score ( ):

[0161] If the waiting time is less than 60 seconds (the first time threshold), according to the stepped gain mechanism, the time gain score is 0 when the waiting time is less than 60 seconds. = 0 points

[0162] Determine the compatibility factor (f(Direction)):

[0163] The elevator is descending on the 8th floor. A person on the 3rd floor requests to descend. The 3rd floor is below the 8th floor and the direction is the same (both are descending). According to Table 1, this is determined to be a "forward reward" scenario, so f(Direction) = 1.5.

[0164] Calculate the marginal cost of stopping ( Distance energy consumption = |3 - 8| × 2 = 5 × 2 = 10 minutes. According to Table 2, the energy consumption for opening and closing the door is 5 minutes (for humans). = 10 + 5 = 15 points

[0165] Calculate the forward distance reward (Bonus_distance): The human request is located ahead of the elevator's direction of travel (forward). Among all forward requests (robot on the 5th floor, human on the 3rd floor), the 3rd floor is farther from the 8th floor (5th floor > 3rd floor), thus not meeting the "closest distance" condition. Therefore, Bonus_distance = 0 points.

[0166] Calculate the final score (S)

[0167] = (100 + 0) ×1.5 - 15 + 0 = 135 points.

[0168] Humans requested a final score: 135 points;

[0169] The rating comparison is shown in Table 3:

[0170] Table 3: Comparison Results of Scores

[0171] Initial priority score 70 points 100 points Humans 30 points higher Time gain score 0 points 0 points same Compatibility coefficient 1.5 1.5 Same (both are in the same direction) Distance energy consumption 6 points 10 points The robot scored 4 points lower (due to being closer). Energy consumption for opening and closing doors 10 points 5 points The robot scored 5 points higher (but took longer). Marginal parking value 16 points 15 points Robot scores 1 point higher Forward Distance Bonus 50 points 0 points The robot scored 50 points higher (because it was closer). Final score 139 points 135 points Robot scored 4 points higher

[0172] Scheduling decision:

[0173] Robot score (139 points) > Human score (135 points); The elevator should prioritize responding to the robot's request on the 5th floor; The elevator should stop at the 5th floor to pick up the robot first, and then continue down to the 3rd floor to pick up the human passenger.

[0174] In this way, although the robot's initial priority score is low, its overall score is high. The elevator stops at the 5th floor to pick up the robot, and then continues down to pick up the passenger on the 3rd floor. This allows the low-priority tasks to be processed side by side without significantly delaying the passenger on the 3rd floor, thus achieving optimal throughput for the entire building.

[0175] Step 4: Based on the scoring of the elevator call request, a shadow dispatcher is used to send a shadow command to the target elevator, which triggers the corresponding stop action of the target elevator.

[0176] In this embodiment, the score of each elevator call request is calculated, and the elevator calls are completed in descending order of score. A shadow dispatcher sends a shadow command to the target elevator. The specific ways in which the shadow command triggers the corresponding stop action of the target elevator are as follows: 1. Physical layer simulation: the external call button level (12V / 24V pulse signal) is simulated through a relay; 2. Link layer simulation: an internal selection command is injected through a CAN bus or RS485 message, so that the elevator group controller recognizes that an internal call has been generated in the car.

[0177] Secondly, after completing one forward movement (either after the elevator has gone down or up), the score for the remaining elevator call requests is recalculated.

[0178] It should be noted that the shadow scheduler is a non-intrusive elevator command control module. It induces the elevator to execute the scheduling intent of the arbitration gateway by simulating the level signals of the elevator's physical buttons or injecting disguised internal / external call messages into the underlying protocol link, without modifying the elevator's original factory logic. Using the shadow scheduler, non-intrusive elevator control can be achieved without modifying the elevator's kernel code.

[0179] Step 5: When the identity attribute of the elevator call request is a Class 4 identity request, the shadow dispatcher sends a shadow instruction to the target elevator and starts the reservation timer. When the reservation timer reaches the preset buffer time and no passenger with the elevator call request is detected, the shadow dispatcher sends a reverse intervention instruction to cancel the corresponding shadow instruction.

[0180] In this embodiment, the preset buffer time is 90 seconds. A facial recognition camera or infrared detector located outside the elevator lobby detects whether the scheduled user has arrived. When the camera or detector detects the user's arrival, the reservation timer stops. If the time taken to stop reaches the preset buffer time, the shadow scheduler sends a reverse intervention command to cancel the corresponding shadow command. If the time taken to stop is less than the preset buffer time, the shadow command is not canceled.

[0181] By setting up a shadow command cancellation mechanism in this way, the situation where the elevator runs empty due to the reservation user not arriving on time can be avoided.

[0182] Example 2

[0183] refer to Figure 2 An elevator dispatching arbitration system for implementing the aforementioned elevator dispatching arbitration method includes the following units:

[0184] Elevator call request receiving unit: used to receive multi-source elevator call requests, assign a corresponding identity fingerprint to each elevator call request and monitor the waiting time of each elevator call request, and then include the identity fingerprint including identity attributes, initial priority attributes, location attributes and physical attributes;

[0185] Elevator call request judgment unit: used to obtain the current load and remaining car space of the target elevator, and determine whether the current load and remaining car space of the target elevator meet the corresponding elevator call request based on the identity attribute and physical attribute; if the current load or remaining car space of the target elevator does not meet the corresponding elevator call request, the elevator call request is cancelled; if the current load and remaining car space of the target elevator both meet the corresponding elevator call request, then proceed to step 3.

[0186] Elevator call request scoring unit: It is used to calculate the elevator call request score based on the initial priority attribute, waiting time and location attribute using the scheduling scoring model;

[0187] Elevator stop control unit: Used to score based on elevator call requests, and uses a shadow dispatcher to send shadow commands to the target elevator, which triggers the corresponding stop action of the target elevator.

[0188] In this embodiment, the specific workflow of the elevator dispatch arbitration system is as follows: the elevator call request receiving unit receives all elevator call requests, assigns a corresponding identity fingerprint to the elevator call request and detects the waiting time of the elevator call request, sends the identity fingerprint of the elevator call request to the elevator call request judgment unit, and sends the waiting time to the elevator call request scoring unit.

[0189] After receiving the identity fingerprint of the elevator call request, the elevator call request judgment unit determines whether the identity attribute of the elevator call request is a third-type identity request. If it does not belong to the third type, the identity fingerprint is sent to the elevator call request scoring unit. If it does belong to the third type, the unit determines whether the current load capacity and remaining car space of the target elevator meet the corresponding elevator call request. If the current load capacity or remaining car space of the target elevator does not meet the corresponding elevator call request, the elevator call request is canceled and a signal is sent to the robot to guide it to move to another available elevator. If the current load capacity and remaining car space of the target elevator both meet the corresponding elevator call request, the identity fingerprint of the elevator call request is sent to the elevator call request scoring unit.

[0190] The elevator call request scoring unit calculates a score for the call request based on the initial priority attribute, waiting time, and location attribute using a scheduling scoring model. This score is then sent to the elevator stop control unit. Based on the call request score, the elevator stop control unit uses a shadow scheduler to send a shadow command to the target elevator, which triggers the corresponding stop action on the target elevator.

[0191] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. An elevator dispatching arbitration method, characterized in that, Includes the following steps: Step 1: Receive multi-source elevator call requests, assign a corresponding identity fingerprint to each elevator call request and monitor the waiting time of each elevator call request, and then include the identity fingerprint in the form of identity attributes, initial priority attributes, location attributes and physical attributes; Step 2: Obtain the current load and remaining car space of the target elevator. Based on the identity attribute and physical attribute, determine whether the current load and remaining car space of the target elevator meet the corresponding call request. If the current load or remaining car space of the target elevator does not meet the corresponding call request, cancel the call request. If the current load and remaining car space of the target elevator both meet the corresponding call request, proceed to Step 3. Step 3: Based on the initial priority attribute, waiting time and location attribute, the scheduling scoring model is used to calculate the elevator call request and obtain the elevator call request score; Step 4: Based on the scoring of the elevator call request, a shadow dispatcher is used to send a shadow command to the target elevator, which triggers the corresponding stop action of the target elevator.

2. The elevator dispatching arbitration method according to claim 1, characterized in that, In step 1, the identity attributes include four types of identity requests: the first type of identity request includes fire-fighting requests and emergency medical requests; the second type of identity request includes facial recognition requests and access control card swipe requests; the third type of identity request includes logistics robot requests and delivery robot requests; and the fourth type of identity request includes indoor unit requests and APP remote reservation requests. The initial priority attribute includes the initial score of the first type of identity request, the initial score of the second type of identity request, the initial score of the third type of identity request, and the initial score of the fourth type of identity request; the initial score of the first type of identity request > the initial score of the second type of identity request > the initial score of the third type of identity request > the initial score of the fourth type of identity request. The location attribute is the floor where the elevator call request is located; The physical properties include weight and geometric dimensions.

3. The elevator dispatching arbitration method according to claim 2, characterized in that, Step 2 includes the following sub-steps: Step A1: Obtain the current load and remaining space in the car of the target elevator; Step A2: Based on the identity attribute, determine whether the elevator call request belongs to the third type of identity request. If the elevator call request does not belong to the third type of identity request, proceed to step 3; if the elevator call request belongs to the third type of identity request, proceed to step A3. Step A3: Set the load threshold and, based on the physical size attribute, determine whether the current load of the target elevator is less than the load threshold and whether the remaining space of the target elevator car is greater than the physical size attribute; if the current load of the target elevator is not less than the load threshold, or the remaining space of the target elevator car is not greater than the physical size attribute, then the stop action is not triggered; if the current load of the target elevator is less than the load threshold and the remaining space of the target elevator car is greater than the physical size attribute, then proceed to step 3.

4. The elevator dispatching arbitration method according to claim 1, characterized in that, Step 3 includes the following sub-steps: Step B1: Obtain the initial score corresponding to the elevator call request based on the initial priority attribute of the elevator call request; Step B2: Based on the waiting time of the elevator call request, a step gain mechanism is used to judge the waiting time and obtain the time gain score of the elevator call request; Step B3: Based on the target elevator status and the location attributes of the floor and the call request, obtain the compatibility coefficient; Step B4: Based on the energy consumption of the distance the target elevator moves to the call request and the energy consumption of the switching time, obtain the marginal stop cost; Step B5: Input the initial score, time gain score, compatibility coefficient, and marginal stop cost into the scheduling score model to obtain the score for the elevator call request.

5. The elevator dispatching arbitration method according to claim 4, characterized in that, In step B2, the step gain mechanism is as follows: When the waiting time for an elevator call request is less than the first time threshold, the time gain score is 0. When the first time threshold is less than or equal to the waiting time for the elevator call request and less than the second time threshold, the time gain score is the first set value. When the second time threshold is less than or equal to the waiting time for the elevator call request and less than the third time threshold, the time gain score is the second set value. When the third time threshold is less than the waiting time for the elevator call request, the time gain score is the third set value.

6. The elevator dispatching arbitration method according to claim 1, characterized in that, The formula for the scheduling scoring model is: Where S is the rating of the elevator call request. The initial score corresponding to the elevator call request. For compatibility coefficient, Value for marginal stops Forward distance reward.

7. The elevator dispatching arbitration method according to claim 2, characterized in that, It also includes step 5: When the identity attribute of the elevator call request is a fourth type of identity request, the shadow dispatcher sends a shadow instruction to the target elevator and starts the reservation timer. When the reservation timer reaches the preset buffer time and no passenger with the elevator call request is detected, the shadow dispatcher sends a reverse intervention instruction to cancel the corresponding shadow instruction.

8. An elevator dispatching arbitration system, characterized in that, A method for implementing the elevator dispatch arbitration method according to any one of claims 1-7 includes the following units: Elevator call request receiving unit: used to receive multi-source elevator call requests, assign a corresponding identity fingerprint to each elevator call request and monitor the waiting time of each elevator call request, and then include the identity fingerprint including identity attributes, initial priority attributes, location attributes and physical attributes; Elevator call request judgment unit: used to obtain the current load and remaining car space of the target elevator, and determine whether the current load and remaining car space of the target elevator meet the corresponding elevator call request based on the identity attribute and physical attribute; if the current load or remaining car space of the target elevator does not meet the corresponding elevator call request, the elevator call request is cancelled; if the current load and remaining car space of the target elevator both meet the corresponding elevator call request, then proceed to step 3. Elevator call request scoring unit: It is used to calculate the elevator call request score based on the initial priority attribute, waiting time and location attribute using the scheduling scoring model; Elevator stop control unit: Used to score based on elevator call requests, and uses a shadow dispatcher to send shadow commands to the target elevator, which triggers the corresponding stop action of the target elevator.

Images