Vehicle dispatch system and vehicle dispatch method

Abstract

The purpose of the present invention is to reduce energy consumption for transporting passengers in a vehicle dispatch system. A method executed by this vehicle dispatch system may include acquiring occupancy data relating to a destination location for a first vehicle. The method may furthermore include determining a predicted occupancy value of the destination location using the occupancy data. The method may furthermore include assessing whether the predicted occupancy value of the destination location satisfies a prescribed occupancy criterion. The method may furthermore include, in response to it being assessed that the predicted occupancy value does not satisfy the prescribed occupancy criterion, determining, using a search process, a second vehicle that can arrive at the destination location earlier than an arrival time of the first vehicle. The method may furthermore include, in response to it being assessed that the predicted occupancy value does not satisfy the prescribed occupancy criterion, automatically transmitting a command to the second vehicle to occupy the destination location.

Description

Vehicle Allocation System and Vehicle Allocation Method

[0001] The present invention relates to a vehicle allocation system and a vehicle allocation method.

[0002] Transportation network data can be collected from sensors within vehicles moving in the transportation network and from infrastructure sensors proximate to the transportation network. For example, transportation network data can be obtained from fixed infrastructure such as traffic cameras and inductive loop traffic sensors, vehicle-mounted sensors, the self-reported location of vehicles, and information from connected road users. Processing the collected transportation network data can not only provide meaningful insights for predicting behavior within the transportation network, but can also be used to control automated vehicles (vehicles) within the transportation network. For example, as described in Patent Document 1, transportation network data can be used to control flying vehicles (aircraft) and arrive at a desired location used by road vehicles.

[0003] Patent Document 1 describes a control device and a control method capable of reliably securing a parking location using a flying vehicle until a passenger transportation vehicle arrives at the parking location. The control device controls the flying vehicle to occupy the parking location and wait until the passenger transportation vehicle arrives at the parking location.

[0004] International Publication No. WO2021 / 005783

[0005] This summary is provided to introduce the selection of concepts that will be further described in the following “Modes for Carrying Out the Invention.” This summary is not intended to identify any important or essential features of the claimed subject matter, nor is it intended to be used to help limit the scope of the claimed subject matter. Transporting passengers using human-driven and autonomous vehicles may require securing stops for passengers to access the vehicle, or destinations where passengers will alight from the vehicle, or both. Depending on the traffic frequency at a particular stop, available stops at the end of a route may already be occupied before the vehicle picks up passengers at the destination. On the other hand, stops may be occupied while transporting passengers along the desired route and before the vehicle reaches its final destination. If there are no available places for passengers to board or alight, the vehicle's route may need to be detoured to another destination away from the desired destination. Detours and unexpected waiting times required to find available stops may result in increased energy consumption and loss of transport time for transporting passengers and other cargo.

[0006] To address these technical challenges, conventional solutions (Patent Document 1) have attempted to secure parking spaces using drones and other flying devices (flying vehicles) before the vehicle's arrival. However, drones and other flying devices often have limitations on their available service areas. For example, flying vehicles may not be permitted in densely populated areas where permission is often required to conduct flight operations. Airspace restrictions may also limit the use of unmanned flying vehicles to assist vehicles in transporting passengers.

[0007] In general, in one embodiment, the embodiment relates to a system including a first vehicle, a second vehicle, a network, and a vehicle dispatch manager connected to the first and second vehicles via the network. The vehicle dispatch manager includes a computer processor and memory. The vehicle dispatch manager obtains occupancy data for a destination location for the first vehicle. The destination location is located at the end of the route the first vehicle is traveling from a departure location. The vehicle dispatch manager uses the occupancy data to determine a predicted occupancy value for the destination location. The predicted occupancy value corresponds to the probability that the first vehicle can occupy the destination location at a given time. The vehicle dispatch manager determines whether the predicted occupancy value of the destination location meets a predetermined occupancy criterion. In response to determining that the predicted occupancy value does not meet the predetermined occupancy criterion, the vehicle dispatch manager uses a search process to determine a second vehicle that can arrive at the destination location earlier than the arrival time of the first vehicle. In response to determining that the predicted occupancy value does not meet the predetermined occupancy criterion, the vehicle dispatch manager automatically sends a command to the second vehicle to occupy the destination location.

[0008] In general, in one embodiment, the embodiment relates to a method comprising a computer processor obtaining occupancy data relating to a destination location for a first vehicle. The destination location is located at the end of a route the first vehicle is traveling on from a departure point. The method further includes the computer processor using the occupancy data to determine a predicted occupancy value for the destination location. The predicted occupancy value corresponds to the probability that the first vehicle can occupy the destination location at a given time. The method further includes the computer processor determining whether the predicted occupancy value of the destination location satisfies a predetermined occupancy criterion. In response to the computer processor determining that the predicted occupancy value does not satisfy the predetermined occupancy criterion, the method further includes using a search process to determine a second vehicle that can arrive at the destination location earlier than the arrival time of the first vehicle. In response to the computer processor determining that the predicted occupancy value does not satisfy the predetermined occupancy criterion, the method further includes the computer processor automatically sending a command to the second vehicle to occupy the destination location.

[0009] In some embodiments, the first vehicle includes a controller, a global positioning system (GPS) device, a seat sensor, and a camera device. Occupancy data may include GPS data from the GPS device, sensor data from the seat sensor, and image data from the camera device. Occupancy data is acquired by various vehicles during various routes for transporting various passengers.

[0010] In some embodiments, the second vehicle includes a controller that performs autonomous driving and occupies the destination location without a human driver until the first vehicle arrives.

[0011] In some embodiments, the vehicle dispatch manager includes a machine learning model. The machine learning model may take as input the departure location of a first vehicle, the destination location of a first vehicle, vehicle demand data for the destination location, and image data of the destination location. The machine learning model may output a predicted occupancy value.

[0012] In some embodiments, a camera device is connected to a vehicle dispatch manager via a network. The camera device can acquire image data of the destination location in real time. Occupancy data may include image data.

[0013] In some embodiments, a user device is connected to a vehicle dispatch manager via a network. The user device can send a reservation request to the vehicle dispatch manager using a graphical user interface. The reservation request may correspond to a requested route for transporting passengers from a first location to a second location. Occupancy data may include passenger usage data relating to the reservation request, the first location, and the second location.

[0014] In some embodiments, the command is a notification to a second vehicle that includes a request for the first vehicle to occupy a destination location. The second vehicle may send a response over the network indicating acceptance of the request to wait at the destination location.

[0015] In some embodiments, a notification is received that the second vehicle is waiting at the destination location. In response to the determination that the first vehicle has arrived at the destination location, a command can be sent to the second vehicle to leave the destination location.

[0016] In some embodiments, the vehicle dispatch manager selects a second vehicle to occupy the destination location after the first vehicle has begun moving toward the destination location.

[0017] In some embodiments, the predicted occupancy value is based on passenger use at the destination location by the passenger's own service vehicle, demand forecast results of user booking requests, in-service usage data based on the travel history of at least one vehicle, and / or real-time video of the destination location acquired by a fixed camera installed at the destination location.

[0018] In some embodiments, the departure timing of the second vehicle is determined relative to the route to the destination based on the predicted occupancy and the travel time of the second vehicle.

[0019] In light of the above structure and function, embodiments of the present invention may include means adapted to perform the various steps and functions defined above according to one or more embodiments and any one of the embodiments of one or more embodiments described herein. Other aspects and advantages of the claimed subject matter will become apparent from the following description and the appended claims.

[0020] Using predictive occupancy data, a vehicle dispatch system can automatically assign a vehicle to occupy a designated location to guarantee a pick-up / drop-off spot for another vehicle. Thus, the vehicle dispatch system can overcome various limitations of unmanned aerial vehicles and other methods for securing parking spaces for vehicles (automobiles). This allows the vehicle dispatch system to reduce energy consumption for transporting passengers through optimization of parking locations and route optimization to reduce waiting times for both passengers picking up and dropping off along the requested route.

[0021] This is a diagram showing a system according to several embodiments. This is a diagram showing a system according to several embodiments. This is a diagram showing a system according to several embodiments. This is a flowchart according to several embodiments. This is a diagram showing an example according to several embodiments. This is a diagram showing an example according to several embodiments. This is a diagram showing an example according to several embodiments. This is a diagram showing a computing system according to several embodiments. This is a diagram showing a computing system according to several embodiments.

[0022] Herein, specific embodiments of the disclosed technology are described in detail with reference to the attached figures. Similar elements in the various figures are indicated by the same reference numerals for consistency. In the following “Modes for Carrying Out the Invention” of the embodiments of this disclosure, numerous specific details are described in order to provide a more complete understanding of this disclosure. However, it will be apparent to those skilled in the art that this disclosure can be carried out without these specific details. In other examples, well-known features are not described in detail to avoid unnecessarily complicating the explanation.

[0023] Throughout this application, ordinal numbers (e.g., 1st, 2nd, 3rd, etc.) may be used as adjectives of elements (i.e., any noun in this application). The use of ordinal numbers does not imply or create a particular order of elements, nor does it limit any element to only a single element, unless expressly disclosed, such as by using the terms “before,” “after,” “single,” and other such terms. Rather, the use of ordinal numbers is for distinguishing elements. For example, the first element is different from the second element, and the first element may encompass two or more elements and follow (or precede) the second element in the order of elements.

[0024] In general, embodiments of the present disclosure include systems and methods for predicting the availability of a stop for passengers to board or alight at a given time. For example, a vehicle dispatch system can automatically determine the predicted occupancy status of a stop using different types of data, such as actual image data of the stop, demand data for that stop, and vehicle usage data collected by vehicles traveling to multiple stops along a route. More specifically, a vehicle dispatch system can analyze a predicted occupancy value of a particular stop against a given occupancy criterion. The predicted occupancy value may indicate the probability that a particular vehicle will be able to occupy the stop, and the given occupancy criterion may be a threshold indicating whether the stop is likely to be occupied by an obstacle (e.g., a parked vehicle or pedestrian) at the time of the vehicle's arrival. Thus, a vehicle dispatch manager can, in response to a user request, determine the likelihood of successfully starting a route and reaching its endpoint.

[0025] Figure 1 is a schematic diagram of a vehicle transport network (e.g., vehicle transport network X (100)) which may include various vehicles (e.g., vehicle A (111), vehicle B (112), vehicle N (113)), user devices (e.g., user device Z (190)), various network elements (not shown), and a vehicle dispatch manager (e.g., vehicle dispatch manager Y (170)). User devices may include personal computers, smartphones, smartwatches, human-machine interfaces, and any other devices connected to the computer network that receive user input from the user. User devices may include input devices, display devices, and hardware and / or software having functionality to provide a graphical user interface (GUI). Network elements may refer to various hardware components in the computer network, such as switches, routers, and hubs, as well as any other logical entities for integrating one or more physical devices on the computer network, such as user devices, servers, network storage devices, user equipment, or the Internet. The vehicle dispatch manager, user devices, controllers, and network elements may include a computing system similar to the computing system (800) described in Figures 8A and 8B below and in the accompanying description.

[0026] Furthermore, the vehicle may include one or more controllers (e.g., controller A (121), controller B (122)), one or more vehicle-mounted sensors (e.g., sensor A (141) in the case of vehicle A, sensor B (142) in the case of vehicle B), a Global Positioning System (GPS) device (e.g., GPS device A (131) in the case of vehicle A, GPS device B (132) in the case of vehicle B), and / or one or more camera devices (e.g., camera device A (151) in the case of vehicle A, camera device B (152) in the case of vehicle B). The vehicle may be an autonomous vehicle, a semi-autonomous vehicle, or a human-driven vehicle. For example, an autonomous vehicle may include an automobile that includes an advanced driver-assistance system (ADAS) that can automate, adapt, and / or enhance the vehicle system for safety and improved driving performance by avoiding obstacles, notifying the driver of potential hazards, and / or correcting driving errors. Autonomous or semi-autonomous vehicles may have the capability to autonomously navigate a portion of a vehicle transport network without human input. For example, a vehicle may be configured to use an autonomous mode in which it navigates through a vehicle transport network with little or no input from a human driver.

[0027] Furthermore, the vehicle may include a controller (e.g., controller A(121) in the case of vehicle A, controller B(122) in the case of vehicle B) which includes hardware and / or software that has the functionality to monitor and / or adjust the vehicle's movement and different vehicle states. Thus, the controller can monitor external objects in the vicinity of the vehicle, such as roadside hazards and nearby vehicles. The controller can receive vehicle data and infrastructure data regarding the vehicle's speed, location, driving status, destination, route (e.g., from route data B(182)), vehicle sensors, and external object data (e.g., object speed and object location). The controller can also perform various autonomous driving functions for the autonomous vehicle.

[0028] Vehicle sensors may include speed sensors, wheel speed sensors, seating sensors, gyroscopes, optical sensors, laser sensors, radar sensors, acoustic sensors, or any other sensor types capable of identifying the state of the vehicle or the vehicle's environment. For example, seating sensors may be mounted on the seats of a vehicle to detect the pressure when an occupant is positioned (seated) in the vehicle's seat. Thus, seating sensors can be used to determine the presence or absence of vehicle occupants and their location (e.g., the driver's seat, front passenger seats, or rear passenger seats). Sensors can also detect external environmental conditions of the vehicle, such as road shape and stationary or moving obstacles such as other vehicles, cyclists, and pedestrians. Examples of vehicle sensors may include hardware sensors such as Light Detection and Ranging (LIDAR) sensors and Radio Detection and Ranging (RADAR) sensors. The vehicle can also identify and track objects approaching it using various detection signals from different sensors.

[0029] The vehicle may further include various camera devices such as video cameras, laser detection systems, infrared detection systems, and acoustic detection systems for recording images and / or video data of the interior and exterior of the vehicle. A camera device (e.g., camera device X(162)) may be placed at a parking location (e.g., parking location X(160), parking location Z(165)). For example, a camera device can collect occupancy data (e.g., occupancy data A(181)) regarding the availability of a parking location for the vehicle to use for passenger boarding and alighting. More specifically, obstacles such as parked vehicles, pedestrians, or road hazards (e.g., obstacle X(161)) may prevent the vehicle from occupying the parking location. Thus, the occupancy data may indicate the presence or absence of obstacles and how often obstacles occur at a given location.

[0030] A GPS device may include various hardware and / or software for acquiring positioning signals from various positioning satellites. For example, a GPS device may include an antenna, receiver, transmitter, processor, and / or one or more tracking loop circuits for use in determining location data. Using the positioning signals, the GPS device can determine location data for a vehicle that indicates the location of the GPS device relative to the positioning satellites. The location data from the GPS device can then be used for the occupation data of each vehicle.

[0031] Furthermore, the vehicle dispatch manager can collect data (e.g., map data Y(173), predicted occupancy data Y(174), route data Y(175), image data Y(176), usage data Y(177), demand data Y(178)) and / or send commands (e.g., command Z(183)) to various vehicles based on the collected data. For example, the vehicle dispatch manager can use commands to implement autonomous driving control. In addition, the vehicle dispatch manager can use commands to guide passenger transport vehicles to specific locations. Thus, a command can indicate a specific location for dropping off passengers, as well as one or more alternative stopping locations if the desired location is already occupied. In this way, the vehicle dispatch manager can send a command to another vehicle to occupy a stopping location before a passenger transport vehicle arrives at that location. Examples of commands include network messages transmitted via machine-to-machine network protocols, or control signals that, upon receipt, automatically trigger the operation of one or more vehicles (e.g., by instructing a controller in an autonomous vehicle). However, the commands may also include notifications to human drivers displayed in the user interface that require a human driver to be present at a specific location (for example, to wait for a passenger transport vehicle).

[0032] Figure 2 shows an exemplary functional block of a vehicle dispatch manager (270). The vehicle dispatch manager (270) may include a server (271) having various processing units such as an already occupied location prediction unit (211), a standby execution determination unit (212), a standby vehicle setting unit (213), a standby vehicle driving instruction unit (214), a standby result receiving unit (215), a departure and arrival database (216), a route determination unit (217), a map database (218), a dispatch plan database (219), and a reservation unit (220). In particular, these functional units can be realized by a hardware processor executing a program. For example, the program that can be executed may be stored in memory or other storage devices, or a program received via a communication interface. Since the various functional units are realized by loading a program, as described above, a part or the whole of one part (function) may be provided in another part. Alternatively, these functions may be realized in hardware by configuring electronic circuits, etc., that realize a part or the whole of each function. The vehicle dispatch manager (270) may also be a controller. For example, a controller can be deployed or distributed across one or more vehicles, one or more servers, and / or one or more devices connected to a vehicle transport network, enabling a variety of functions.

[0033] Continuing to refer to Figure 2, the occupied location prediction unit (211) can acquire occupied data regarding destination locations for a particular vehicle. Occupation data may be acquired by various vehicles communicating with the vehicle dispatch manager (270) between various routes for transporting various passengers. The occupied data may also be image data of destination locations acquired in real time by a camera device. For example, a camera device can transmit image data to the vehicle dispatch manager (270) via the vehicle transport network while the vehicle is traveling along a route. In addition, destination locations for a route can be acquired from a departure and arrival database (216) (which stores automatically determined locations, such as destination and departure locations requested by the user, as well as locations collected from software applications on the user device while traveling along the route). Destination locations may also be determined by a route determination unit (217), which determines the route between two locations and the departure location from map data (for example, from a map database (218)). The occupied location prediction unit (211) can use the occupied data to further determine the predicted occupied value of the destination location. Furthermore, the predicted occupancy value may correspond to the probability that a vehicle can occupy the destination location at a predetermined time. The previously occupied location prediction unit (211) can further determine whether the predicted occupancy value of the destination location meets a predetermined occupancy criterion.

[0034] Furthermore, in response to the determination that the predicted occupancy value does not meet a predetermined occupancy criterion, the standby execution determination unit (212) can use a search process to determine a standby vehicle that can arrive at the destination location earlier than the arrival time of the passenger transport vehicle. Similarly, in response to the determination that the predicted occupancy value does not meet a predetermined occupancy criterion, the standby vehicle setting unit (213) can automatically send a command to the standby vehicle to occupy the destination location. Based on the command, the selected standby vehicle can send a response that is received by the standby result receiving unit (215) on the vehicle dispatch manager (270). The result may be a message from the human driver of the potential standby vehicle indicating acceptance or rejection of the standby request. Alternatively, the result may be an automated acknowledgment that the controller in the selected vehicle will or cannot perform the requested action. Standby vehicles can be selected by obtaining vehicle data from the dispatch plan database (219). The standby vehicle setting unit (213) can also obtain notification that one vehicle is waiting at the destination location for another vehicle. The standby vehicle setting unit (213) can, in response to determining that a passenger transport vehicle has arrived at its destination, send a further command to the standby vehicle to leave the destination. The standby vehicle setting unit (213) can select a standby vehicle to occupy the destination after the passenger transport vehicle has begun moving toward its destination.

[0035] Furthermore, the standby vehicle driving instruction unit (214) can transmit the determination result of the occupancy of a parking space to the unassigned vehicle. The standby implementation determination unit (212) can receive the determination result, which includes route data and arrival time information for occupying the parking space, from the destination arrival determination unit (for example, the destination arrival determination unit (322) of the vehicle (311) described below). The standby vehicle setting unit (213) can send a notification to the passenger vehicle that the parking space has been secured, based on the already assigned vehicle that is occupying the parking space. In this way, the already assigned vehicle can proceed towards the parking space on the planned route until the passenger vehicle's arrival time.

[0036] After receiving notification, the waiting vehicle can yield its position to a passenger transport vehicle at the boarding / alighting point and proceed to a new location, such as to wait for future dispatches. In this way, the destination arrival determination unit can calculate the destination or route for the undispatched vehicle to return to the waiting point. At the new location, the undispatched vehicle can wait until another passenger vehicle requires the new parking space to be occupied. For example, the waiting vehicle travel instruction unit (214) can obtain the arrival time of the passenger vehicle to the boarding / alighting point from the dispatch plan database (219), while also setting a destination or route in the vehicle control unit of the already dispatched vehicle so that it arrives at the parking space at or before the arrival time.

[0037] If a dispatched vehicle cannot stop at the boarding point, the standby implementation determination unit (212) can send a notification to the passenger vehicle stating that a waiting vehicle cannot secure a parking space for the passenger. In this way, the standby vehicle setting unit (213) can change the passenger vehicle's boarding point to another location (for example, a location closer than the original boarding point). The vehicle dispatch manager (270) can check whether there is demand for other vehicles to dispatch a vehicle other than the original passenger transport vehicle. The vehicle dispatch manager can return the dispatched vehicle to its original position to wait for future dispatches. Due to the determination that a waiting vehicle cannot secure a parking space, a notification is sent to the passenger vehicle before the passenger vehicle arrives at the desired parking location. This allows the passenger vehicle to stop at a location closer to the destination on the desired route than the destination location, thereby reducing energy consumption. The vehicle dispatch manager can also check whether there is demand for other vehicles to arrive at the parking location later than the passenger vehicle. If such vehicle demand exists, the waiting vehicle can receive instructions from the waiting vehicle dispatch instruction unit (214) and continue to secure the desired parking location. If there is no vehicle demand for the desired parking location, the vehicle dispatch manager can terminate all attempts to secure the desired parking location and return the dispatched vehicle to its original undispatched position.

[0038] Furthermore, the reservation unit (220) within the vehicle dispatch manager (270) can receive reservation requests from user devices requesting transportation for one or more passengers to their destinations. User devices can submit reservation requests using a graphical user interface, and the reservation request corresponds to the requested route for transporting passengers between locations. Thus, the occupancy data may include passenger usage data related to the reservation request and the locations in the reservation request. The route determination unit (217) can determine the timing of the vehicle's departure to the destination location based on the predicted occupancy value and the vehicle's travel time.

[0039] The vehicle dispatch manager may include hardware and / or software having functionality for generating and / or updating machine learning models (e.g., machine learning model Y(171), machine learning model (231)). For example, a machine learning model may take as input the vehicle's departure location, the vehicle's destination location, vehicle demand data for the destination location, and image data of the destination location. These inputs may be used by the machine learning model to output predicted occupancy values ​​for parking locations. In addition, various types of machine learning models can be trained, such as convolutional neural networks, deep neural networks, recurrent neural networks, support vector machines, decision trees, inductive learning models, deductive learning models, supervised learning models, unsupervised learning models, and reinforcement learning models. In a deep neural network, for example, layers of neurons can be trained with a predetermined list of features based on the outputs of preceding network layers. Thus, as data passes through the deep neural network, more complex features may be identified in the data by neurons in subsequent layers.

[0040] To train a model, various types of machine learning algorithms can be used, such as the backpropagation algorithm (e.g., machine learning algorithm Y(172), machine learning algorithm (232)). In the backpropagation algorithm, the gradient is calculated in reverse for each hidden layer of the neural network, starting from the layer closest to the output layer and proceeding to the layer closest to the input layer. Thus, the gradient can be calculated using the transpose of the weights of each hidden layer based on an error function (also called a "loss function"). The error function can be based on various criteria, such as the mean squared error function or a similarity function, and the error function can be used as a feedback mechanism to adjust the weights in the machine learning model (e.g., one of the machine learning models Y(171)).

[0041] Regarding artificial neural networks, for example, an artificial neural network may include one or more hidden layers, each containing one or more neurons. Neurons can be modeling nodes or objects that roughly mimic neurons in the human brain. In particular, neurons can combine data inputs with a set of coefficients, i.e., a set of network weights to adjust the data inputs. Through machine learning, a neural network can determine which data inputs should be given higher priority when determining one or more specified outputs of the artificial neural network. Regarding recurrent neural networks (RNNs), a recurrent neural network can repeatedly perform a specific task on a large number of data elements in an input sequence, and the output of the recurrent neural network depends on past calculations. Thus, a recurrent neural network can operate in memory or hidden cell states that provide information for use by the current cell calculations for the current data input.

[0042] Figure 3 shows an exemplary functional block of a vehicle (311) including a controller (320), a powertrain (330), a user interface (312), a display device (313), a steering system (314), a chassis (315), a camera device (316), sensors (317), and wheels (318). The controller (320) includes a vehicle control unit (321), a destination arrival determination unit (322), a memory (323), a processor (324), and a communication interface (325). The vehicle control unit (321) can perform route planning, navigation, and driving control. For example, the vehicle control unit (321) can be used to move a portion of a vehicle transport network according to one or more predetermined vehicle operations, such as parking the vehicle, starting the engine, maintaining lane and vehicle separation, and operating peripheral devices (e.g., headlights or interior lights). More specifically, the vehicle control unit (321) can determine various command values ​​for steering, driving, and braking based on profile information (e.g., desired route, target vehicle speed, etc.) from the destination arrival determination unit (322).

[0043] In addition, the destination arrival determination unit (322) can determine the arrival time for the vehicle (311) to complete a specific route between the departure point and the destination. For example, the destination arrival determination unit (322) can determine the vehicle's travel route from a starting point, such as the vehicle's current location or the departure point after passengers have been picked up, to the destination. The travel route can be based on vehicle information (e.g., past driver history), environmental information (e.g., weather conditions), and vehicle traffic network information (e.g., current vehicle traffic on one or more roads along the route), or a combination thereof. Similarly, the vehicle control unit (321) can control the vehicle (311) to move through the vehicle transport network according to the route. The autonomous vehicle can output the travel route to a trajectory controller so that the autonomous vehicle can be driven from the departure point to the destination using the specific route.

[0044] Furthermore, the powertrain (330) includes a power source (331), a vehicle actuator (332), a transmission (333), a drive shaft (334), and other vehicle components such as a suspension, an axle, and / or an exhaust system. Thus, the powertrain (330) can supply kinetic energy and / or electric power to various components within the vehicle (311). For example, the controller (320) can receive electric power from the powertrain (330) and communicate with the powertrain (330) and / or the wheels (318) to control the vehicle (311), which control can include controlling the vehicle (311) in terms of acceleration, deceleration, steering, or other aspects. The power source (331) can be any device capable of supplying energy such as electrical energy, thermal energy, or kinetic energy. In particular, the power source (331) may be an engine such as an electric motor or an internal combustion engine that can provide motive power to various wheels. The power source (331) can include potential energy units such as lithium-ion batteries, lead-acid batteries, and deep cycle batteries, as well as other energy sources such as solar cells. The transmission (333) can receive energy from the power source (331) that is relayed to individual wheels to generate movement of the vehicle. The vehicle actuator (332) can receive signals from the controller (320) to operate the power source (331), the transmission (333), and other vehicle components.

[0045] Figures 1, 2, and 3 show various configurations of components, and other configurations may be used without departing from the scope of the present disclosure. For example, various components in Figures 1, 2, and 3 can be combined to create a single component. As another example, functionality performed by a single component may be performed by two or more components.

[0046] FIG. 4 is a diagram illustrating a general method for determining predicted occupancy data and / or dispatching waiting vehicles based on the predicted occupancy data. The blocks in FIG. 4 can be executed by various components (e.g., vehicle dispatching manager Y (170)) as described in FIGS. 1, 2, and 3. Although the various blocks in FIG. 4 are presented and described sequentially, those skilled in the art will understand that some or all of the blocks may be executed in a different order, combined, or omitted, and some or all of the blocks may be executed in parallel. Further, the blocks may be executed actively or passively.

[0047] In block 400, a request to transport one or more passengers using a passenger vehicle is obtained. For example, a user can use a user device to send a reservation request to the vehicle dispatching manager via a network. The vehicle dispatching manager can match a vehicle for transporting one or more passengers along the requested route with the user. Similarly, the reservation request may simply be a request for a vehicle, in which case the desired route is determined in real time by the passengers who enter the vehicle.

[0048] In block 405, the departure location and / or destination location of the passenger vehicle for the route is obtained. For example, the user device can provide the departure location and / or destination location via user input to the user device. Thereafter, the user device, the passenger vehicle, and / or the vehicle dispatching manager can determine the route between the departure location and the destination location. For example, map data can be obtained from a map database to determine various via points and other trip information for generating the route. Based on the determined route, the vehicle dispatching manager can determine the arrival time of the passenger vehicle at the departure location and / or destination location along the route. The departure location and / or destination location can be obtained from a database storing the routes assigned to various vehicles within the vehicle transportation network.

[0049] In block 410, the arrival times of one or more passenger vehicles for the departure and / or destination locations are determined.

[0050] In block 415, occupancy data relating to the departure and / or destination locations is obtained. For example, the occupancy data may indicate the frequency and duration for which a particular stop can be used as either a departure or destination location for transporting passengers. The occupancy data may also be binary, such as whether or not a passenger vehicle is likely to occupy a stop on a particular route. On the other hand, the occupancy data may also indicate a number of data attributes. For example, one data attribute may identify the total number of vehicles that may occupy a stop at a particular time (for example, if the stop is a parking lot, the total number of vehicles may correspond to the number of empty parking spaces). Other data attributes may include various types of availability (for example, whether an occupied vehicle can be permanently parked at the stop, or whether an occupied vehicle can only be present for a limited period, such as 15 minutes).

[0051] Occupancy data may include vehicle data acquired from various vehicles (e.g., passenger-only service vehicles, ride-sharing vehicles, and autonomous vehicles) for one or more parking locations. For example, a vehicle may automatically record occupancy data during a past route indicating whether the vehicle could occupy a particular location. Occupancy data may be based on different types of vehicle data, such as seating data from seating sensors (e.g., whether the vehicle experienced a change in the number of passengers at a particular location), driving data (e.g., whether the vehicle was able to park at that location for a predetermined amount of time or whether the vehicle entered a parking operation at a particular time on a recorded route), and / or image data from cameras (e.g., image data can identify the number of parked vehicles and / or available parking locations while traveling along a particular route).

[0052] Occupancy data may further include passenger usage data for specific stops. For example, a vehicle dispatch manager may record whether the passenger's journey to a specific departure and / or destination location was successful. Similarly, occupancy data may include whether a waiting vehicle was needed to complete the route for transporting the passenger. Passenger usage data can be collected from vehicle data recorded by the vehicle dispatch manager. Similarly, passenger usage data may also be collected from various user devices (e.g., GPS coordinates of the passenger along the desired route). Similarly, passenger usage data may include frequency data, such as how often a user requests a route from a departure and / or destination location. Thus, passenger usage data can identify various periods of low and high occupancy frequency for stops.

[0053] A vehicle dispatch manager can determine demand data for a specific stop based on passenger usage at locations along a desired route. For example, a vehicle dispatch manager can predict the number of passengers arriving at, boarding, or alighting at a stop. Occupancy data may also include in-service usage data from vehicles on various routes. Vehicles can also automatically acquire stopping records of different vehicles at a specific location via the vehicle transport network. Dashboard cameras can also record obstacles at a stop over different periods while on other routes. In this way, occupancy data can be used to determine statistical data regarding the availability of a stop.

[0054] Occupancy data can be collected in real time for various parking locations. For example, a fixed camera device can be installed near a specific location to collect real-time image data. Vehicle detection sensors, such as overhead indicator sensors, ground sensors, and surface-mounted sensors, can also be placed at parking locations. The vehicle detection sensors and fixed camera devices can then transmit their occupation data to the vehicle dispatch manager via the vehicle transport network.

[0055] In block 420, one or more predicted occupancy values ​​for departure and / or destination locations are determined using occupancy data and multiple arrival times. For example, the predicted occupancy values ​​can identify the likelihood that a parking location is occupied at a particular arrival time. The predicted occupancy values ​​can be determined using statistical analysis and / or rule-based algorithms (e.g., a lookup table that matches occupancy data with each occupancy value).

[0056] Predicted occupancy values ​​can also be determined using one or more machine learning models. In particular, machine learning models can be trained on historical occupancy data (e.g., historical image data of various locations, historical demand data of those locations, and historical vehicle usage data). The machine learning model can acquire various inputs, such as historical and real-time data, to predict occupancy values ​​before a vehicle completes its route to a parking location. The machine learning model can also use occupancy data for similar locations at a given geographical distance from the location of interest. The machine learning model may be an artificial neural network, such as a recurrent neural network.

[0057] In block 430, a determination is made as to whether the predicted occupancy value meets a predetermined occupancy criterion. For example, the predetermined occupancy criterion may correspond to whether the predicted occupancy status of a desired stopping location is above a certain threshold. If the predicted occupancy value is below the threshold, the vehicle dispatch manager can determine that the vehicle will fail to stop at the particular location. The vehicle dispatch manager can also determine, based on the predicted occupancy value, whether the passenger route can be completed without using one or more waiting vehicles based on the predetermined occupancy criterion.

[0058] Figure 5 shows predetermined values ​​for thresholds used as predetermined occupancy criteria based on the occupancy rate of a stopping location. In Figure 5, the number of vehicles α currently present at a pick-up / drop-off location can be determined using occupancy data (e.g., using in-service usage data and fixed camera data). The vehicle dispatch manager can then determine the number of vehicles β that will be added to the pick-up / drop-off location in the future to be used as a predicted occupancy value. The predicted occupancy value can then be corrected for the pick-up / drop-off location by monitoring passenger usage data at the location.

[0059] The predetermined occupancy criteria can be dynamic values ​​that are periodically updated based on passenger usage data and / or passenger demand data at the stop. Stops that are frequently occupied may require predetermined occupancy criteria that make it more likely for passenger vehicles to occupy the stop than places that are less frequently occupied. The predetermined occupancy criteria may also include the requirement that the stop can be occupied before the vehicle begins to move. If it is determined that the predicted occupancy values ​​of the departure and / or destination locations meet the predetermined occupancy criteria, the process can proceed to block 460. If it is determined that the departure and / or destination locations do not meet the predetermined occupancy criteria, the process can proceed to block 440.

[0060] In block 440, a search process is used to determine one or more routes and one or more arrival times for various vehicles. Using the current location data of available vehicles (e.g., vehicles located within a given geographical proximity or on non-passenger transport routes that do not overlap), the vehicle dispatch manager can determine various routes to a particular stop. In this way, the search process can determine which waiting vehicles are available to occupy a stop by comparing distance, route, and other factors. The search process may also be an iterative process that analyzes available candidates in communication with the vehicle dispatch manager. This stop may correspond to the departure or destination location of a passenger vehicle's route. Based on the determined routes, the arrival times of available vehicles can also be determined. Thus, the vehicle dispatch manager can determine whether a particular vehicle is likely to arrive at the stop before the passenger vehicle arrives. Using the arrival times, the vehicle dispatch manager can also determine whether an available vehicle can occupy the stop for a predetermined amount of time before the passenger vehicle arrives. In this way, the vehicle dispatch manager can also use predicted occupancy data when analyzing the arrival times of available vehicles.

[0061] In block 445, one or more waiting vehicles are determined from a variety of searched vehicles based on one or more routes and one or more arrival times for the departure and / or destination locations. Based on the determined routes and arrival times, the vehicle dispatch manager can select waiting vehicles from the available vehicles. In addition to using routes and arrival times, the vehicle dispatch manager may also use other factors to select a particular vehicle, such as fuel efficiency and whether available vehicles are already carrying passengers to the stop.

[0062] In block 450, one or more commands are sent to one or more waiting vehicles to occupy a departure and / or destination location. After selecting the waiting vehicles, the vehicle dispatch manager can send a command to the selected vehicles to occupy a parking location. Thus, the command can cause the controller in the autonomous vehicle to automatically reroute to the requested waiting location. In the case of a human-driven vehicle, the command can generate a request in the display device that requests the human driver to reroute. In the case of a human-driven vehicle, the driver can accept the request and proceed to the parking location.

[0063] In block 455, one or more waiting vehicles are used to occupy the departure and / or destination locations based on one or more commands.

[0064] In block 460, one or more passengers are transported from the departure point to the destination point using a passenger vehicle.

[0065] Figure 6 shows an example of vehicle dispatch, where the user is transported to their destination in response to boarding a vehicle following a reservation request. As shown in Figure 6, there is a high probability that another vehicle may be parked at the destination before the user's arrival. Accordingly, a waiting vehicle can be dispatched to the destination to ensure a available space for the user to disembark at the destination.

[0066] Figure 7 illustrates another example of vehicle dispatch based on vehicle reservations for two different users and a waiting vehicle to occupy a shared parking space. As shown in the figure, User B's departure location partially overlaps in time with User A's destination location. Therefore, the vehicle dispatch manager can adjust the waiting vehicle to occupy the parking space until User B's vehicle arrives. The waiting vehicle can then depart to operate on a completely different route, while User A's vehicle can occupy the parking space immediately after User B's vehicle departs.

[0067] Figure 8A shows a computing system (800) which may include any combination of mobile, desktop, server, router, switch, embedded device, or other types of hardware. The computing system (800) may include one or more computer processors (802), non-persistent storage (804) (e.g., volatile memory such as random access memory (RAM) and cache memory), persistent storage (806) (e.g., optical drives such as hard disks, compact disk (CD) drives or digital versatile disk (DVD) drives, flash memory, etc.), and communication interfaces (812) (e.g., Bluetooth interface, infrared interface, network interface, optical interface, etc.). The computer processor (802) may be an integrated circuit for processing instructions. For example, the computer processor may be one or more cores or microcores of a processor. The computing system (800) may also include one or more input devices (810), such as a touchscreen, keyboard, mouse, microphone, touchpad, electronic pen, or any other type of input device. The communication interface (812) may include an integrated circuit for connecting the computing system (800) to a network (for example, a local area network (LAN) or a wide area network (WAN) such as the Internet or a cellular network).

[0068] Furthermore, the computing system (800) may include one or more output devices (808), such as a screen (e.g., a liquid crystal display (LCD), a plasma display, a touchscreen, a projector, or other display device), a printer, external storage, or any other output device. One or more of the output devices may be the same as or different from the input devices. The computing system (800) may implement and / or connect to a data repository. For example, one type of data repository is a database. A database is a collection of information configured to facilitate the retrieval, modification, reorganization, and deletion of data. A database management system (DBMS) is a software application that provides an interface for users to define, create, query, update, or manage databases.

[0069] Software instructions, in the form of computer-readable program code for performing various functions, may be stored, in whole or in part, temporarily or permanently, on non-temporary computer-readable media such as storage devices, diskettes, tapes, flash memory, physical memory, or any other computer-readable storage medium. Specifically, software instructions may correspond to computer-readable program code configured to perform various functions when executed by a processor.

[0070] The computing system (800) in Figure 8A may be connected to a network or be part of a network. For example, as shown in Figure 8B, the network (820) may include a number of nodes (e.g., node X (822), node Y (824)). Each node may correspond to a computing system, such as the computing system shown in Figure 8A, or a group of combined nodes may correspond to the computing system shown in Figure 8A. Nodes in the network (820) (e.g., node X (822), node Y (824)) may be configured to provide services to a client device (826). Nodes may include functionality to receive requests from the client device (826) and send responses to the client device (826). The client device (826) may be a computing system, such as the computing system shown in Figure 8A.

[0071] The computing system (800) may be implemented as part of a cloud computing system. For example, a node may be part of a cloud computing system. For example, a cloud computing system may include remote servers along with various other cloud components such as cloud storage units and edge servers. Thus, a cloud computing system may have different functions distributed across numerous locations from a central server, which can run using an internet connection.

[0072] A computing system may further include the functionality to receive data from a user. For example, a user may submit data through a graphical user interface (GUI) on their user device. Data can be submitted through the GUI by the user selecting one or more GUI components, or by inserting text and other data into GUI widgets using a touchpad, keyboard, mouse, or any other input device. In response to the selection of a particular item, information about that item may be retrieved by the computer processor from persistent or non-persistent storage. When an item is selected by the user, the content of the retrieved data about that item may be displayed on the user device in response to the user's selection.

[0073] Although only a few exemplary embodiments have been described in detail above, those skilled in the art will readily understand that many modifications are possible in the exemplary embodiments without substantially departing from the present invention. Accordingly, all such modifications are intended to be within the scope of this disclosure as defined in the appended claims.

Claims

1. A vehicle dispatch system comprising: a first vehicle; a second vehicle; and a vehicle dispatch manager connected to the first vehicle and the second vehicle via a network, wherein the vehicle dispatch manager comprises a computer processor, and the computer processor performs the following processes: acquiring occupancy data relating to a destination location located at the end of a route the first vehicle is traveling on from a departure point; using the occupancy data, determining a predicted occupancy value of the destination location corresponding to the probability that the first vehicle can occupy the destination location at a predetermined time; determining whether the predicted occupancy value of the destination location satisfies a predetermined occupancy criterion; in response to the determination that the predicted occupancy value does not satisfy the predetermined occupancy criterion, using a search process, determining a second vehicle that can arrive at the destination location earlier than the arrival time of the first vehicle; and in response to the determination that the predicted occupancy value does not satisfy the predetermined occupancy criterion, automatically sending a first command to the second vehicle to occupy the destination location.

2. The vehicle dispatch system according to claim 1, wherein the first vehicle includes a controller, a Global Positioning System (GPS) device, a seat sensor, and a camera device, the occupancy data includes GPS data from the GPS device, sensor data from the seat sensor, and image data from the camera device, and the occupancy data is acquired by multiple vehicles between multiple routes for transporting multiple passengers.

3. The vehicle dispatch system according to claim 1, wherein the second vehicle comprises a controller configured to perform autonomous driving and occupy the destination location without a human driver until the arrival of the first vehicle.

4. The vehicle dispatch system according to claim 1, wherein the vehicle dispatch manager includes a machine learning model, the machine learning model acquires, as input, the departure location of the first vehicle, the destination location of the first vehicle, vehicle demand data for the destination location, and image data of the destination location, and the machine learning model outputs the predicted occupancy value.

5. The vehicle dispatch system according to claim 1, further comprising a camera device connected to the vehicle dispatch manager via the network, wherein the camera device acquires image data of the destination location in real time, and the occupancy data includes the image data.

6. The vehicle dispatch system according to claim 1, further comprising a user device connected to the vehicle dispatch manager via the network, wherein the user device transmits a reservation request to the vehicle dispatch manager using a graphical user interface, the reservation request corresponds to a requested route for transporting a passenger from a first location to a second location, and the occupancy data includes passenger usage data relating to the reservation request, the first location, and the second location.

7. The vehicle dispatch system according to claim 1, wherein the first command is a notification to the second vehicle including a request for the first vehicle to occupy the destination location, and the second vehicle is configured to transmit a response via the network indicating acceptance of the request to wait at the destination location.

8. The vehicle dispatch system according to claim 1, further comprising: processing performed by the computer processor receiving notification that the second vehicle is waiting at the destination location; and, in response to determining that the first vehicle has arrived at the destination location, sending a second command to the second vehicle to leave the destination location.

9. The vehicle dispatch system according to claim 1, wherein the vehicle dispatch manager selects the second vehicle to occupy the destination location after the first vehicle has started moving toward the destination location.

10. The vehicle dispatch system according to claim 1, wherein the predicted occupancy value is based on at least one of the following: passenger use at the destination location by the passenger's own service vehicle; demand forecast results of reservation requests by users; service usage data based on the driving history of at least one vehicle; and real-time video of the destination location acquired by a fixed camera installed at the destination location.

11. The vehicle dispatch system according to claim 1, further comprising the process performed by the computer processor determining the departure timing of the second vehicle for the route to the destination location based on the predicted occupancy value and the travel time of the second vehicle.

12. A vehicle dispatch method comprising: a computer processor acquiring occupancy data relating to a destination location located at the end of a route taken by a first vehicle from a departure point; the computer processor using the occupancy data to determine a predicted occupancy value of the destination location corresponding to the probability that the first vehicle can occupy the destination location at a predetermined time; the computer processor determining whether the predicted occupancy value of the destination location satisfies a predetermined occupancy criterion; the computer processor, in response to the determination that the predicted occupancy value does not satisfy the predetermined occupancy criterion, using a search process to determine a second vehicle that can arrive at the destination location earlier than the arrival time of the first vehicle; and the computer processor, in response to the determination that the predicted occupancy value does not satisfy the predetermined occupancy criterion, automatically sending a first command to the second vehicle to occupy the destination location.

Images