Method, computer program, device, vehicle and network component for controlling maneuvers within a platoon

Abstract

Embodiments provide methods, computer programs, apparatus, vehicles, and network components for controlling maneuvers within a platoon of multiple vehicles. A method (10) for controlling maneuvers within a platoon of multiple vehicles (100; 102) includes: receiving (12) information related to a maneuver of the platoon; determining (14) information about a fuel efficiency of the maneuver based on fuel consumption of maneuvering from an initial state to a maneuvering target state, maintaining the maneuvering target state, and restoring from the maneuvering target state to the initial state; and deciding (16) whether to perform the maneuver based on the information about the fuel efficiency.

Description

Technical Field

[0001] The present invention relates to methods, computer programs, apparatus, vehicles and network components for controlling maneuvers within a platoon of multiple vehicles and more particularly, but not exclusively, to determining whether a platoon maneuver is justified by considering the overall fuel efficiency of the maneuver. Background Art

[0002] Vehicle communications are an area of ​​research and development. To enable autonomous or semi-autonomous driving of vehicles, it is desirable for vehicles to use vehicle-to-vehicle communications (V2V) and vehicle-to-network (V2N) communications, for example, to coordinate driving maneuvers and / or receive remotely operated driving instructions. The communications are typically wireless, i.e., vehicles may communicate wirelessly with other vehicles in their vicinity and / or with backend services via cellular mobile communication systems.

[0003] Automatic or autonomous driving is also an area of ​​research and development. One concept for dealing with high traffic loads is platooning, where vehicles are grouped and can allow for more efficient use of road capacity, lower fuel consumption, respectively. Groups of vehicles (also known as convoys or platoons) can be used to take advantage of the short distance or headway between vehicles to operate the vehicles in the platoon, since the vehicles in the platoon can react within a short time delay or react simultaneously. This can be achieved through effective control mechanisms between the vehicles in the platoon.

[0004] Within the concept of Agile Quality of Service Adaptation (AQoSA), the applications supported by the communication system adapt their settings to a foreseeable Quality of Service (QoS). This is particularly important when the application in question is a safety-related, time-critical application such as High Density Platooning (HDPL) where the Inter-Vehicle Distance (IVD) is below 15 m in order to benefit from reduced drag and, thereby, lower fuel consumption. At this IVD, the sensor system may need to be supported by information transmitted by other vehicles, since the required detection and reaction times are very short. The quality of the communication link is therefore crucial, since the performance of the application depends to a large extent on it. For example, vehicles reduce their IVD in order to reduce their fuel consumption. They can achieve this distance reduction due to good communication conditions, about which they may already know in advance (Predictive Quality of Service, PQoS). However, when the QoS degrades, the IVD must be increased again.

[0005] Document DE 10 2017 204 326 A1 provides further details on quality of service monitoring in mobile communication systems, in particular for vehicle applications, such as for platoons. Document DE 103 56 256 A1 describes a concept for quality monitoring of a radio interface. Document DE 10 2012 212 339 A1 discloses a grouping concept for vehicles. Document DE 10 2016 226 050 A1 describes a concept for pre-allocating or reserving radio resources required in the future in order to ensure a certain quality of service. Document DE 10 2016 006 523 A1 discloses a concept for setting the distance between vehicles based on a specific traffic situation.

[0006] LIANG KUO-YUN et al., “Heavy-Duty Vehicle Platoon Formation for FuelEfficiency” (IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, IEEE, PISCATAWAY, NJ, USA, vol. 17, no. 4, April 1, 2016 (2016-04-01), pp. 1051-1061) studied how two or more dispersed vehicles can cooperate to form a platoon in a fuel efficient manner.

[0007] FENGQIAO LUO et al., “Coordinated platooning with multiple speeds” (TRANSPORTATION RESEARCH. PART C, EMERGING TECHNOLOGIES, vol. 90, May 1, 2018 (2018-05-01), pp. 213-225) proposed a coordinated platooning model with multiple speed options, which integrates scheduling, routing, speed selection, and platoon formation / disbandment.

[0008] "When is it Fuel Efficient for a Heavy DutyVehicle to Catch Up With a Platoon" by KUO-YUN LIANG et al. ” (7TH IFAC SYMPOSIUM ON ADVANCES INAUTOMOTIVE CONTROL, January 1, 2013 (2013-01-01), pp. 738-743) examined the following question: When is it beneficial for a heavy vehicle to accelerate in order to catch up and join a platoon?

[0009] JORNOD GUILLAUME et al., "Packet Inter-Reception Time Modeling for High-Density Platooning in Varying Surrounding Traffic Density" (EUROPEAN CONFERENCE ON NETWORKS AND COMMUNICATIONS (EUCNC), IEEE, June 18, 2019 (2019-06-18, pp. 187-192) deals with agile quality of service adaptation, where applications and communication systems exchange quality of service requirements and predictions.

[0010] An improved concept for controlling platoon maneuvers is needed.

[0011] Embodiments are based on the discovery that the overall fuel efficiency of a platoon maneuver can be determined based on the fuel efficiency of the maneuver target state and based on the fuel efficiency of a transition maneuver to and from the maneuver target state. Another discovery is that for transition maneuvers, there is a tradeoff between the fuel efficiency of the transition maneuver and the overall fuel efficiency. For example, if the maneuver is to reduce IVD to a certain value, then the IVD can be reduced by performing a deceleration maneuver on the vehicle in front until the IVD is reached. The IVD can be increased by performing a deceleration maneuver on the vehicle in the rear of the platoon. Deceleration maneuvers are more fuel efficient than acceleration maneuvers. However, the overall way progress of the platoon is reduced by the deceleration maneuver, which is also contemplated by the embodiments. Summary of the invention

[0012] Embodiments provide a method for controlling maneuvers within a platoon of multiple vehicles. The method includes: receiving information related to maneuvers of the platoon. The method further includes: determining information about the fuel efficiency of the maneuver based on fuel consumption of maneuvering from an initial state to a maneuvering target state; maintaining the maneuvering target state; and reverting from the maneuvering target state to the initial state. The maneuvering further includes: deciding whether to perform the maneuver based on the information about the fuel efficiency. Embodiments provide for efficient maneuvering control in a platoon through overall fuel efficiency evaluation.

[0013] The maneuver may be a high density platoon maneuver for which the inter-vehicle distance for the maneuver target state is determined based on the communication latency between the vehicles. The maneuver may be controlled relative to a given time frame (e.g., a predicted time frame) and a target state (e.g., a minimum IVD given by the communication latency). Fuel inefficient maneuvers may be avoided, and transition maneuvers may be improved or even optimized in terms of the overall fuel efficiency of the maneuver.

[0014] In some embodiments, the method further comprises: determining an inter-vehicle distance for the maneuvering target state based on a predicted quality of service of inter-vehicle communications in the platoon. Embodiments may further adapt, improve, or optimize the target state based on the communication quality of the inter-vehicle communications.

[0015] Information about fuel efficiency may be further based on: the overall duration of maneuvering from the initial state to the maneuver target state, holding / maintaining the maneuver target state, and resuming from the maneuver target state to the initial state. Subdividing the maneuver in transition maneuvers to and from the maneuver target state may enable further improvements in fuel efficiency and assessment of whether the overall fuel efficiency of the maneuver justifies the maneuver as such.

[0016] In some embodiments, the determination may include determining an effective time of the maneuvering target state. The effective time may be based on the maneuver from the initial state to the maneuvering target state, the maintenance of the maneuvering target state, and the recovery from the maneuvering target state to the initial state. Embodiments may use the effective time as an efficient model for evaluating and adapting the maneuver.

[0017] The determination may further include determining information about fuel efficiency based on the effective time of the maneuvering target state and the fuel saving rate during the maneuvering target state.Then, in an embodiment, the effective time may model the overall maneuver according to the predicted quality of service.

[0018] For example, determining the effective time of the maneuvering target state may include: determining the actual time of maneuvering from the initial state to the target state and the compensation time of maneuvering from the initial state to the target state. Determining the effective time of the maneuvering target state may further include: determining the actual time of recovering from the target state to the initial state and the compensation time of recovering from the target state to the initial state. In an embodiment, the compensation time can be used to model the positive and negative contributions of the transition maneuver to the overall maneuver in an efficient manner.

[0019] In some embodiments, the method may further include: evaluating different effective times of different reference points for maneuvers from an initial state to a target state and for recovery from a target state to an initial state. A reference point may indicate a vehicle in the row that forms a maneuvering reference for other vehicles in the row. For example, if the last vehicle in the row forms a maneuvering reference, all other vehicles may slow down to reduce the IVD in the row. Slowing down is more fuel efficient, but the overall driving distance of the row is lost. If the leading vehicle in the row forms a reference, all other vehicles may accelerate to reduce the IVD in the row. Similarly, if the last vehicle is a reference and the IVD is to be increased, all vehicles except the last vehicle will accelerate, and the driving distance may increase. Acceleration is less fuel efficient, but the overall driving distance of the row may increase. Embodiments may consider using different reference vehicles for different transition maneuvers, and thereby improving overall fuel efficiency.

[0020] Different effective times may result in different actual times and different compensation times, wherein at least one of the compensation times is negative. Thus, some maneuvers (e.g., acceleration maneuvers with the last vehicle as a reference) may be considered to have negative compensation times because the overall travel distance of the platoon may be increased.

[0021] Vehicles of the plurality of vehicles may communicate with each other using a mobile communication system. The overall duration of the maneuver may be determined by a predicted quality of service in the mobile communication system. Such prediction may be performed at the vehicle, at an infrastructure component of the mobile communication system, or may involve both. For example, statistical information about the quality of service may be obtained and the prediction may be based on the statistical information. For example, the overall duration of the maneuver may be determined based on the predicted quality of service. The overall duration may enable an overall fuel efficiency assessment of the maneuver.

[0022] An embodiment further provides a computer program having a program code for executing one or more of the methods described above when the computer program is executed on a computer, a processor or a programmable hardware component. A further embodiment is a computer-readable storage medium storing instructions which, when executed by a computer, a processor or a programmable hardware component, causes the computer to implement one of the methods described herein.

[0023] Embodiments further provide an apparatus for controlling maneuvers within a platoon of multiple vehicles. The apparatus comprises one or more interfaces for communicating with one or more vehicles in the platoon of vehicles. The apparatus further comprises a control module configured to perform one of the methods described herein.

[0024] A vehicle comprising an embodiment of the device and a network component comprising an embodiment of the device are further embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Some other features or aspects will be described, by way of example only and with reference to the accompanying drawings, the following non-limiting embodiments of using an apparatus or a method or a computer program or a computer program product, in which:

[0026] Figure 1 A block diagram illustrating an embodiment of a method for controlling maneuvers within a platoon is illustrated;

[0027] Figure 2 illustrates an embodiment of an apparatus for controlling maneuvers within a platoon and a block diagram of an embodiment of a vehicle;

[0028] Figure 3 The diagram illustrates the concept of distance investment in an embodiment;

[0029] FIG4 illustrates scene snapshots: (a) a simulation model of a test track facility, and (b) a zoom-in on a five-truck platoon in an embodiment;

[0030] FIG5 illustrates reference placement in an embodiment for: (a) fuel consumption optimization with reference at the back, (b) maneuver duration optimization with reference at the center, compared to (c) conventional reference at the front, during an IVD closing maneuver;

[0031] Figure 6 illustrates row input parameters and time targets in an embodiment;

[0032] Figure 7 Graph deviation as a function of time and maneuver duration T in an embodiment are shown m Export of

[0033] Figure 8 illustrates the air drag ratio as a function of IVD for five trucks in a platoon in an embodiment;

[0034] Fig. 9 Graphs illustrate relative fuel savings as a function of final IVD for various initial IVDs in an embodiment;

[0035] FIG10 illustrates performance evaluation without drag correction for (a) a closing gap maneuver and (b) an opening gap maneuver for different parameter combinations in an embodiment;

[0036] FIG11 illustrates performance evaluation of (a) a closing gap maneuver and (b) an opening gap maneuver with drag correction for different parameter combinations in an embodiment;

[0037] FIG. 12 illustrates selected results of extended embodiment simulations: for fixed initial IVD for (a) anterior reference, (b) intermediate reference, and (c) posterior reference d i = 30 m of closing maneuver; and (d) for a fixed final distance d f =5m intermediate reference opening maneuver;

[0038] Fig.13 is an illustrative representation of an IVD during transition between in-line to HDPL and back from HDPL to in-line in an embodiment;

[0039] Fig.14 A diagram showing the effect of a reference combination on a final distance budget in an embodiment;

[0040] FIG. 15 illustrates a surface model created by graphically combining the duration of a maneuver and relative fuel consumption in an embodiment using a Lasso regression model; and

[0041] Fig.16 The embodiment for d f = 5m fuel saving percentage as a function of favourable time. DETAILED DESCRIPTION

[0042] Various embodiments will now be described more fully with reference to the accompanying drawings in which some embodiments are illustrated. In the accompanying drawings, the thickness of lines, layers or regions may be exaggerated for clarity. Broken lines, dashed lines or dotted lines may be used to illustrate optional components.

[0043] Accordingly, although the exemplary embodiments are capable of various modifications and alternative forms, their embodiments are shown in the figures by way of example and will be described in detail herein. However, it should be understood that there is no intention to limit the exemplary embodiments to the specific forms disclosed, but on the contrary, the exemplary embodiments are intended to cover all modifications, equivalents and alternatives that fall within the scope of protection of the present invention. The same reference numerals refer to the same or similar elements throughout the description of the figures.

[0044] As used herein, the term "or" refers to a non-exclusive "or" unless otherwise indicated (e.g., "or else," or "or alternatively"). In addition, as used herein, words used to describe the relationship between elements should be broadly interpreted to include direct relationships or the presence of intermediate elements, unless otherwise indicated. For example, when an element is referred to as being "connected" or "coupled" to other elements, the element may be directly connected or coupled to the other elements or there may be intermediate elements. In contrast, when an element is referred to as being "directly connected" or "directly coupled" to another element, there are no intermediate elements. Similarly, words such as "between," "adjacent," etc. should be interpreted in a similar manner.

[0045] The terms used herein are only for the purpose of describing specific embodiments and are not intended to limit exemplary embodiments. As used herein, the singular forms "one", "an" and "the" are also intended to include plural forms, unless the context clearly indicates otherwise. It will be further understood that the terms "include", "comprise", "include", "include" or "include" when used herein specify the presence of stated features, integers, steps, operations, elements or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components or groups thereof.

[0046] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as those commonly understood by those skilled in the art to which the exemplary embodiments belong. It will be further understood that terms (e.g., terms defined in commonly used dictionaries) should be interpreted as having a meaning consistent with their meaning in the context of the relevant art, and will not be interpreted in an idealized or overly formal sense unless explicitly defined as such herein.

[0047] Figure 1 A block diagram of an embodiment of a method for controlling a maneuver within a platoon is illustrated. The method 10 for controlling a maneuver within a platoon of a plurality of vehicles includes: receiving 12 information related to a maneuver of the platoon. The method 10 further includes: determining 14 information about a fuel efficiency of the maneuver based on a fuel consumption of maneuvering from an initial state to a maneuvering target state; maintaining the maneuvering target state; and restoring from the maneuvering target state to the initial state. The method 10 further includes: deciding 16 whether to perform the maneuver based on the information about the fuel efficiency.

[0048] Examples of maneuvers for the platoon are an opening maneuver to add another vehicle to the platoon; a closing maneuver to close a gap of a leaving vehicle, thereby reducing the IVD to a target IVD within a certain time, etc. A platoon is a group or fleet of vehicles, such as coordinated motor vehicles, cars, trucks, etc. For example, the vehicles can communicate with each other directly or using the infrastructure of a mobile communication system. For example, the vehicles in the platoon travel together on a common road section and travel with reduced IVD (fuel saving). At a certain point, a maneuver may be necessary or possible, and in at least some embodiments it may be determined whether such a maneuver is beneficial. An embodiment determines fuel efficiency, for example, it may be assessed whether the overall fuel consumption is higher or lower if a certain maneuver is performed. In addition, a maneuver or a portion thereof may be modified to make the maneuver more fuel efficient. The fuel efficiency (i.e., whether the platoon consumes more or less fuel with the maneuver) may then form the basis for deciding whether to perform the maneuver. .

[0049] Figure 2 A block diagram of an embodiment of an apparatus 20 for controlling maneuvers within a platoon and an embodiment of a vehicle 100 is illustrated. The apparatus 20 includes at least one (one or more) interface 22 for communicating with one or more vehicles in a platoon of multiple vehicles. The apparatus 20 includes a control module 24 coupled to the at least one interface 22. The control module 24 can be configured to control the one or more interfaces 22 and perform any of the methods described herein. Figure 2 An embodiment of a vehicle / network assembly 100 including an embodiment of the apparatus 20 is further illustrated.

[0050] The device 20, the vehicle 100 and the network component 100 may communicate via the mobile communication system 400. Figure 2 As shown, the mobile communication system 400 may correspond, for example, to one of the third generation partnership project (3GPP) standardized mobile communication networks, wherein the terms mobile communication system and mobile communication network are used synonymously. Thus, information related to the motor vehicle may be obtained via the mobile communication system 400, for example, from Figure 2 In other embodiments, the request may be received from a central entity (network component) that controls the vehicles 100, 102 at least to some extent. Likewise, information related to the decision of whether to perform the maneuver may be provided to the vehicle 102 or the network component.

[0051] The mobile communication system 400 may correspond to one of the 3rd Generation Partnership Project (3GPP) standardized mobile communication networks, wherein the term "mobile communication system" is used synonymously with "mobile communication network." The mobile or wireless communication system 400 may correspond to a fifth generation (5G, or new radio) mobile communication system and may use millimeter wave technology. The mobile communication system may correspond to or include, for example, Long Term Evolution (LTE), Advanced LTE (LTE-A), High Speed ​​Packet Access (HSPA), Universal Mobile Telecommunications System (UMTS) or UMTS Terrestrial Radio Access Network (UTRAN), Evolved UTRAN (e-UTRAN), Global System for Mobile Communications (GSM) or Enhanced Data Rates for GSM Evolution (EDGE) network, GSM / EDGE Radio Access Network (GERAN), or a mobile communication network with a different standard, such as the Worldwide Interoperability for Microwave Access (WIMAX) network IEEE 802.16 or the Wireless Local Area Network (WLAN) IEEE 802.11, generally an Orthogonal Frequency Division Multiple Access (OFDMA) network, a Time Division Multiple Access (TDMA) network, a Code Division Multiple Access (CDMA) network, a Wideband CDMA (WCDMA) network, a Frequency Division Multiple Access (FDMA) network, a Space Division Multiple Access (SDMA) network, and the like.

[0052] Service provision may be performed by a network component 100, such as a base transceiver, a relay station, or a UE, for example, to coordinate service provision in a cluster or group of multiple UEs / vehicles. The base transceiver may be operable or configured to communicate with one or more active mobile transceivers / vehicles 100, and the base transceiver may be located in or adjacent to the coverage area of ​​another base transceiver, such as a macrocell base transceiver or a small cell base transceiver. Thus, an embodiment may provide a mobile communication system 400 comprising two or more mobile transceivers / vehicles 100 and one or more base transceivers, wherein the base transceiver may establish a macrocell or a small cell, such as a microcell, a metropolitan cell, or a femtocell. A mobile transceiver or UE may correspond to a smart phone, a cellular phone, a laptop, a notebook computer, a personal computer, a personal digital assistant (PDA), a universal serial bus (USB) stick, a car, a vehicle, and the like. A mobile transceiver may also be referred to as a user equipment (UE) or a mobile device conforming to 3GPP terminology. The vehicle may correspond to any conceivable means of transport, such as a car, a bicycle, a motorcycle, a van, a truck, a bus, a ship, a boat, an airplane, a train, a tram, etc.

[0053] The base transceiver station may be located in a fixed or stationary part of a network or system. The base transceiver station may be or correspond to a remote radio head, a transmission point, an access point, a macro cell, a small cell, a micro cell, a femto cell, a metropolitan area cell, etc. The base transceiver station may be a wireless interface of a wired network that enables the transmission of radio signals to a UE or a mobile transceiver. Such radio signals may comply with radio signals as standardized by, for example, 3GPP or generally conform to one or more of the systems listed above. Therefore, the base transceiver station may correspond to a NodeB, an eNodeB, a gNodeB, a base transceiver station (BTS), an access point, a remote radio head, a relay station, a transmission point, etc., which may be further subdivided into remote units and central units.

[0054] The mobile transceiver or vehicle 100, 102 may be associated with a base transceiver station or a cell. The term "cell" refers to the coverage area of ​​a radio service provided by a base transceiver station (e.g., NodeB (NB), eNodeB (eNB), gNodeB, remote radio head, transmission point, etc.). A base transceiver station may operate one or more cells on one or more frequency layers, and in some embodiments, a cell may correspond to a sector. For example, a sector may be implemented using a sector antenna that provides characteristics for covering an angular portion around a remote unit or base transceiver station. In some embodiments, a base transceiver station may, for example, operate three or six cells that cover a 120° sector (in the case of three cells), a 60° sector (in the case of six cells), respectively. A base transceiver station may operate multiple sectorized antennas. In the following, a cell may refer to a corresponding base transceiver station that generates the cell, or similarly, a base transceiver station may refer to a cell generated by the base transceiver station.

[0055] In an embodiment, the device 20 may be included in a vehicle, a base station, a NodeB, a UE, a relay station or any service coordination network entity. It should be noted that the term network component may include multiple subcomponents, such as a base station, a server, etc. A further embodiment is a vehicle 100 including the device 20 and / or a network component including the device 20.

[0056] In an embodiment, one or more interfaces 22 may correspond to any component for obtaining, receiving, transmitting or providing analog or digital signals or information, such as any connector, contact, pin, register, input port, output port, conductor, channel, etc., which allows providing or obtaining signals or information. The interface may be wireless or wired, and it may be configured to communicate with other internal or external components, i.e., transmit or receive signals, information. One or more interfaces 22 may include additional components to implement corresponding communications in the mobile communication system 400, which may include transceiver (transmitter and / or receiver) components, such as one or more low noise amplifiers (LNA), one or more power amplifiers (PA), one or more duplexers, one or more antenna duplexers, one or more filters or filter circuits, one or more converters, one or more mixers, correspondingly adapted radio frequency components, etc. One or more interfaces 22 may be coupled to one or more antennas, which may correspond to any transmitting and / or receiving antennas, such as horn antennas, dipole antennas, patch antennas, sector antennas, etc. The antennas may be arranged in defined geometric arrangements, such as uniform arrays, linear arrays, circular arrays, triangular arrays, uniform field antennas, field arrays, combinations thereof, etc. In some examples, one or more interfaces 22 may be used for the purpose of transmitting or receiving information, or both, such as information related to capabilities, application requirements, trigger indications, requests, message interface configurations, feedback, information related to control commands, QoS requirements, QoS maps, etc.

[0057] like Figure 2 As shown, the corresponding one or more interfaces 22 are coupled to the corresponding control module 24 at the device 20. In an embodiment, the control module 24 can be implemented using one or more processing units, one or more processing devices, any components for processing (such as, processors, computers, or programmable hardware components operable with corresponding adapted software). In other words, the described functions of the control module 24 can also be implemented with software, which is then executed on one or more programmable hardware components. Such hardware components may include general-purpose processors, digital signal processors (DSPs), microcontrollers, and the like.

[0058] Figure 2Also shown is an embodiment of a system 400, which includes an embodiment of a UE / vehicle / network component 100 and another vehicle 102, which may also include one or more network components / base stations, which network components / base stations include further embodiments of the apparatus 20. In an embodiment, communication (i.e., transmission, reception, or both) may occur directly among the mobile transceiver / vehicle 100, 102, and / or between the mobile transceiver / vehicle 100, 102 and a network component (infrastructure or mobile transceiver, such as a base station, a network server, a backend server, etc.). Such communication may utilize the mobile communication system 400. Such communication may be performed directly, for example by means of device-to-device (D2D) communication, which in the case of the vehicles 100, 102 may also include vehicle-to-vehicle (V2V) or car-to-car communication. Such communication may be performed using the specifications of the mobile communication system 400.

[0059] In an embodiment, one or more interfaces 22 may be configured to perform wireless communication in the mobile communication system 400. To do so, radio resources are used, such as frequency, time, code and / or space resources, which may be used for wireless communication with a base transceiver station and for direct communication. The assignment of radio resources, i.e., determining which resources are used for D2D and which resources are not used for D2D, may be controlled by the base transceiver station. Here and in the following, the radio resources of the respective components may correspond to any radio resources that may be thought of on a radio carrier, and they may use the same or different granularity on the respective carriers. The radio resources may correspond to a resource block (such as an RB in LTE / LTE-A / unlicensed LTE (LTE-U)), one or more carriers, subcarriers, one or more radio frames, radio subframes, radio slots, one or more code sequences potentially with corresponding spreading factors, one or more spatial resources (such as, spatial subchannels, spatial precoding vectors), any combination thereof, and the like. For example, in direct cellular vehicle-to-anything (C-V2X) (where V2X includes at least V2V, V2-Infrastructure (V2I), etc.), transmissions from 3GPP Release 14 onwards can be managed by the infrastructure (so-called Mode 3) or run in the UE.

[0060] like Figure 2As further illustrated in , the method 10 may be performed at a device 20 in a vehicle 100. The vehicle 100 may receive information related to a maneuver of the platoon, for example, from another vehicle 102, from a control network component, from a platoon determination control entity, etc. The vehicle or network component 100 may then determine 14 information about the fuel efficiency of the maneuver based on the fuel consumption of maneuvering from an initial state to a maneuvering target state, maintaining the maneuvering target state, and returning from the maneuvering target state to the initial state. The vehicle or network component 100 may then decide 16 whether to perform the maneuver based on the information about the fuel efficiency.

[0061] In the following, more details will be provided for some embodiments. In order to cope with changes in the quality of service, AQoSA provides information about the future quality of the link. This information may be accompanied by a prediction horizon, which is the incremental time ending in the future (duration from now) for which the prediction value applies. In some embodiments, knowing the predicted value over time, the cost of the quality of service, the application can calculate how long it will take to use QoS to compensate for the maneuvering cost and service cost.

[0062] A predicted quality of service (PQoS) time series is a sequence of values ​​that represent the future quality of a link. For example, a maneuver may include: a closing maneuver (from the initial inter-vehicle distance (IVD) of the platoon) d i (e.g. 30m) to a smaller final IVD d f Maintain the final IVD d f for a period of time; and then perform an opening maneuver (return to d i ). The group of maneuvers is subject to the length of the favorable QoS period T F (the duration of future QoS during which the platoon can drive with low IVD).

[0063] In another embodiment, the maneuver is a high density platooning (HDPL) maneuver for which the inter-vehicle distance of the maneuver target state is determined based on the communication latency between vehicles. The inter-vehicle distance of the maneuver target state is determined based on the predicted quality of service for inter-vehicle communications in the platoon. The minimum IVD is determined based on the predicted latency for message exchanges, such as based on an emergency braking maneuver. The IVD needs to provide sufficient time buffer for the communication of emergency braking messages.

[0064] Vehicles in the plurality of vehicles communicate with each other using a mobile communication system, such as a mobile communication system that enables V2V or C2C. The maneuver is determined by a predicted quality of service (PQoS) in the mobile communication system. For example, the time range of the PQoS determines the overall duration of the maneuver. Therefore, in the present embodiment, the overall duration of the maneuver is determined based on the PQoS.

[0065] Given a PQoS time series or a PQoS with a time range, the fuel savings for HDPL can be improved or (in theory) maximized. For example, IVD can be reduced as quickly as possible to reach the low air resistance area and benefit from HDPL for the longest time.

[0066] The fuel savings may then be determined from the fuel savings rate and the time of the HDPL.

[0067] Fuel saving = fuel saving rate * HDPL time.

[0068] However, performing closing and opening maneuvers requires a fuel investment (changing IVD requires increasing / decreasing speed and therefore changing fuel consumption). Embodiments may take this finding into account when selecting a maneuvering strategy.

[0069] The information about fuel efficiency can be further based on the total duration of maneuvering from the initial state to the maneuvering target state, maintaining the maneuvering target state, and recovering from the maneuvering target state to the initial state. Determining 14 can include determining the effective time of the maneuvering target state. The effective time can be based on maneuvering from the initial state to the maneuvering target state, maintaining the maneuvering target state, and recovering from the maneuvering target state to the initial state.

[0070] Embodiments can convert relative fuel savings ( F S ) is modeled as the effective HDPL time ( T H ) and the fuel saving rate at the target distance ( ) function.

[0071] .

[0072] Therefore, determining 14 includes: based on the effective time of the maneuvering target state ( F S ) and the fuel saving rate during the maneuvering target state ( ) to determine information about fuel efficiency.

[0073] This concept of effective HDPL time is the HDPL time that takes into account the maneuvering input in fuel and distance. For this purpose, a compensation time for maneuvering fuel input can be introduced.T C .

[0074]

[0075] Where C is the input of one of the closing maneuver and the opening maneuver (transition maneuver). Determining the effective time of the maneuver target state 14 may include: determining the actual time of maneuvering from the initial state to the target state and the compensation time of maneuvering from the initial state to the target state. Similarly, determining the effective time of the maneuver target state 14 may include determining the actual time of returning from the target state to the initial state and the compensation time of returning from the target state to the initial state.

[0076] At least some embodiments may evaluate different validity times for different reference points for maneuvering from an initial state to a target state and for recovery from a target state to an initial state. A reference point indicates a vehicle in the platoon that forms a maneuvering reference for other vehicles in the platoon, such as in Figure 3 Different effective times result in different actual times and different compensation times, where at least one of the compensation times is negative. The time may be negative if the maneuver has saved fuel (e.g., all trucks slowed down to reduce IVD).

[0077] When closing and opening maneuvers create a range deficit, the concept of range investment comes into play. Figure 3 The concept of distance investment in the embodiment is illustrated. Figure 3 A row of three vehicles is shown for which, at a first instant in time, it is shown on the far left, the IVD is d i . Figure 3 The first row of the matrix representation depicts the progression without maneuvers. As can be seen in the middle and on the far right, the IVD for the three vehicles remains constant. Starting from the second row, the development over time for the HDPL maneuver is shown. The IVD decreases to 0.05 in the closed maneuver. d f , and the result is depicted in the middle. Then, IVD increases back to d i , which is shown on the right. Different reference points r c , r o is considered for both closing and opening maneuvers. A value of 0 means that the lead vehicle is the reference point, a value of 0.5 indicates that the center vehicle is the reference point, and a value of 1 indicates that the last vehicle is the reference point. r c , r oThe combinations are shown on the far left. In the second row, the lead vehicle is the reference for both transition maneuvers. As can be seen on the far right, after the opening maneuver, the platoon has made the same progress as in the first row without the maneuver. The position of the first vehicle in this case is marked with the baseline. As can be seen by these combinations, some maneuver configurations increase road progress, while some maneuver configurations lose road progress.

[0078] Thus, the baseline represents a situation where no maneuvers are performed. On the left, the initial state is shown, and in the middle, the HDPL (maneuver target) state is shown. On the right, the final state is shown. Some maneuvers introduce a range loss (the first truck / vehicle behind the baseline (left)) while some maneuvers introduce a range gain (the first truck / vehicle in front of the baseline (right)) compared to the baseline.

[0079] In the case of a loss, the platoon needs to drive a little longer to compensate for the baseline, which in an embodiment can be used to calculate relative fuel savings. In the case of a gain, this means that the platoon then needs to drive less, which means that this can translate into a fuel gain compared to the baseline. r c , r o More details can be found in application EP 19 170 887.4.

[0080] This can be restored to , It is the distance compensation time that takes into account the distance loss or gain of maneuvering.

[0081] In summary, in the embodiment, the effective HDPL time T H It can be obtained in the following ways:

[0082]

[0083] in T F It is the length that is beneficial to PQoS.

[0084] Accordingly, is the maneuver time of the closing maneuver (from the initial state IVD = d i To maneuvering target state IVD = d f ),

[0085] is the fuel input compensation time for closing maneuvers,

[0086] is the maneuver time for opening maneuver (from maneuver target state IVD = d f To the initial state IVD = d i ),

[0087] is the compensation time for the opening maneuver.

[0088] These times are obtained, for example, as described in application EP 19 170 887.4. These times are the final / target IVD d i , maneuver reference, maximum speed deviation and other control parameters (acceleration deviation, control gain, etc.).

[0089] As a result, at least some embodiments may maximize or improve fuel savings for HDPL by addressing the following issues:

[0090]

[0091] in r c is the reference for the closing maneuver, r o is the reference for the opening maneuver, Δ v c is the maximum speed deviation for closing maneuvers, Δ v o is the maximum speed deviation for the opening maneuver, and d f is the final IVD. These parameters can be selected so as to maximize or at least improve fuel savings. d m It is the minimum IVD allowed by PQoS.

[0092] Embodiments may maximize or improve fuel savings by considering compensation time for fuel investment and distance loss / gain for closing and opening maneuvers.In embodiments, HDPL parameters may be selected accordingly.

[0093] Signaling may be implemented in a centralized or distributed manner. For example, a vehicle within the platoon may execute method 10 and notify / coordinate platoon members. In other embodiments, method 10 may be executed at a control center in an infrastructure network component and the results may be transmitted to members of the platoon or coordinating members of the platoon. In further embodiments, a fully distributed implementation is contemplated, wherein method 10 is at least partially executed at each platoon member. Additional coordination or reconcilement communications may then be performed among members within the platoon to ensure group maneuverability.

[0094] As already mentioned, in an embodiment, the corresponding methods may be implemented as computer programs or codes, which can be executed on corresponding hardware. Therefore, another embodiment is a computer program with program code for carrying out at least one of the above methods when the computer program is executed on a computer, a processor or a programmable hardware component. A further embodiment is a computer-readable storage medium storing instructions, which, when executed by a computer, a processor or a programmable hardware component, causes the computer to implement one of the methods described herein.

[0095] The embodiments can be summarized as follows.

[0096] Within the scope of cooperative driving, a promising application is high-density platooning. One of the main goals of this application is to reduce fuel consumption by benefiting from the reduced air drag caused by inter-vehicle distances below ten meters. In order to ensure safety at such low inter-vehicle distances, the application strongly relies on vehicle-to-vehicle communication. To support this type of application, a recent concept is agile quality of service adaptation, where the application receives predictions of the future quality of service from, for example, the network, and adapts its settings to cope with its changes. An important parameter in this concept is the prediction horizon, i.e. the length of the provided predicted quality of service time series. This parameter drives the duration of the reduced air drag benefit as a function of the maneuver time series. In fact, short prediction horizons can be used, but require a fast reaction of the application. This is usually linked to a higher fuel consumption during the adaptation maneuver, thus weakening the original goal of high-density platooning.

[0097] The effect of prediction range on fuel savings is a function of different maneuver parameters. We first propose three strategies to achieve changes in inter-vehicle distance. In an embodiment, the relationship between maneuver duration and fuel consumption for increasing and decreasing inter-vehicle distance can be considered. The maneuver fuel investment can be linked to a compensation time during which the platoon will counterbalance the fuel consumption by benefiting from reduced air drag. Some embodiments may use an optimization method for maximizing fuel efficiency based on some predicted quality of service (PQoS) parameter. In order to benefit from high-density platooning, the prediction range required for a platoon of five trucks may be on the order of a hundred seconds at a minimum.

[0098] An interesting and promising application of cooperative driving is high-density platooning (HDPL). To reduce their fuel consumption, vehicles in HDPL drive with a small inter-vehicle distance (IVD) - 15m, 10m or even 5m. In fact, this reduction can be achieved due to reduced air resistance, see A. Al Alam, A. Gattami and KH Johansson, "An experimental study on the fuel reduction potential of heavy duty vehicle platooning" (13th Int. IEEE Conf. on Intell. Transp. Syst. (ITSC), 2010. IEEE, 2010, pp. 306-311).

[0099] In recent years, truck platooning for energy efficiency has received a lot of attention in the field of cooperative vehicle automation research, see S. Tsugawa, S. Jeschke, and S. E. Shladove, “A review of truck platooning projects for energy savings” (IEEE Trans. on Intell. Veh., vol. 1, no. 1, pp. 68–77, March 2016) and S. van de Hoef, KH Johansson, and DVDimarogonas, “Fuel-efficient en route formation of truck platoons” (IEEE Trans. on Intell. Transp. Syst., vol. 19, no. 1, pp. 102–112, January 2018).

[0100] To achieve this efficiency while ensuring safety, the application requires information exchange with low latency and high reliability. Coordination between vehicles is typically supported by vehicle-to-vehicle (V2V) or vehicle-to-anything (V2X) communications.

[0101] Safety-related, time-critical applications are often limited by the lower bound of the quality of service (QoS) of their communication systems. This can be avoided by applying a recent concept called Agile Quality of Service Adaptation (AQoSA) in communication systems, see A. Pfadler, G. Jornod, A. El Assaad, and P. Jung, "Predictive Quality of Service: Adaptation of Inter Vehicle Distance to Packet Inter-reception Time for HDPL" (IEEE 91st Veh. Technol. Conf. (VTC Spring). IEEE, 2020, under review).

[0102] Applications and communication systems periodically exchange QoS requirements and PQoS, respectively. On the application side, the adaptation of QoS requirements operates together with the adaptation of its functional settings.

[0103] Applied to HDPL, AQoSA allows the platoon to plan its maneuvers. This includes reducing its IVD, maintaining a small IVD, and increasing the IVD. This planning strongly relies on the QoS provided by the communication system. The value in the PQoS time series drives the minimum IVD, which affects the fuel savings. Its length also affects the fuel savings, since the longer the favorable time, the greater the fuel savings. To achieve fuel savings, the duration of this favorable QoS should be longer than a threshold. Depending on the strategy chosen for the application, the requirements for QoS may vary, especially in terms of the forecast horizon.

[0104] The prediction horizon is defined as the total length of the PQoS time series. Thus, in an HDPL maneuver, it may cover the IVD reduction (closing) maneuver, the reduced air drag period, and the IVD increase (opening) maneuver and some compensation time.

[0105] The challenges of PQoS in this range are twofold: (i) the application needs to select an efficient policy based on the provided PQoS time series; (ii) the efficiency of the policy can only be evaluated when the PQoS time series is long enough. To address these challenges, methods for efficient policy conception can be used in embodiments. In addition, the requirements for the prediction range can be precisely derived.

[0106] In an embodiment, the relationship between two objectives of the system and the control strategy can be considered, namely the relationship between fuel consumption and maneuvering time. To this end, an embodiment can provide concepts for group control, relative reference for maneuvers, and other related control parameters. The reference concept can extend the graph-based Laplace control algorithm for closing and opening maneuvers, which is proposed in "Distributed graph-based control of convoys of heterogeneous vehicles using curvilinearroad coordinates" by I. Navarro, F. Zimmermann, M. Vasic, and A. Martinoli (19th IEEE Int. Conf. Intell. Transp. Sys. (ITSC), 2016, pp. 879–886). In the following simulation, the simulation results will be presented, which uses SUMO,

[0107] P. A. Lopez, M. Behrisch, L. Bieker-Walz, J. Erdmann, Y.-P. Flötteröd, R. Hilbrich, L. Lücken, J. Rummel, P. Wagner, and E. Wießner, "Microscopic traffic simulation using SUMO" (21st IEEE Int. Conf. Intell. Transp. Syst. IEEE, 2018) and

[0108] GF Riley and TR Henderson, The ns-3 Network Simulator Using IEEE 802.11p Radio Access Technology, Berlin, Heidelberg: Springer Berlin Heidelberg, 2010, pp. 15–34,

[0109] "Intelligent Transport Systems (ITS); Performance evaluation of selforganizing TDMA as medium access control method applied to ITS; Accesslayer part", European Telecommunications Standards Institute, Tech. Rep. ETSITR 102 862 V1.1.1, 2011.

[0110] The simulated truck platoon can obtain PQoS information from the prediction algorithm running on any node.

[0111] The impact of various external factors on the performance of HDPL supported by different radio access technologies has been studied. These factors are: the number of surrounding nodes and the distance between antennas (see G. Jornod, T. Nan, M. Schweins, A. ElAssaad, A. Kwoczek, and T. Kürner, “Sidelink technologies comparison for highway high-density platoon emergency braking,” 2018 IEEE 16th Int. Conf. Intell. Transp. Syst. Telecommun. (ITST), October 2018, pp. 1–7), and the Doppler shift originating from the incoming traffic (see G. Jornod, R. Alieiev, A. Kwoczek, and T. Kürner, “Environment-aware communications for cooperative collision avoidance applications,” 2018 IEEE 19th Int. Symp. “A World of Wireless, Mobile and Multimedia Networks” (WoWMoM), June 2018, pp. 588–599; and R. Alieiev, G. Jornod, T. Hehn, A. Kwoczek and T. Kürner, “Improving the performance of high-density platooning using vehiclesensorbased doppler-compensation algorithms,” IEEE Transactions on IntelligentTransportation Systems, pp. 1–12, 2019).

[0112] These studies can promote the development of adaptive strategies for QoS changes. The prediction algorithm that provides PQoS can run on the base station, or it can run on the node, such as G. Jornod, A. El Assaad, A. Kwoczek, and T. Kürner, "Packet interreception time modeling for high-density platooning in varying surrounding traffic density" (28th IEEE Eur. Conf. Net. Commun. IEEE, 2019, accepted).

[0113] The fuel consumption model of SUMO can be adapted to reflect the air drag reduction. A Pareto frontierlike relationship between the two objectives can be derived. Using the air drag corrected consumption model, the fuel consumption input can be converted into compensation time. The resulting set of maneuvering time / compensation time for the two maneuvers allows fuel saving optimization or improvement to be performed based on the available PQoS. An embodiment can extend the distributed graph-based feedback fleet controller for closing and opening maneuvers; the relationship between fuel consumption and maneuvering time is considered according to the control strategy, and the fuel savings is improved or optimized according to the PQoS.

[0114] In the following, the scenarios considered in the embodiments and the control models used will be introduced. The method used to evaluate the relationship between fuel consumption and maneuvering time for HDPL in some embodiments will be described in detail. Numerical results will be presented and discussed. In an embodiment, the required prediction range for a given time series can be derived.

[0115] Embodiments may provide a fuel efficient approach for HDPL using PQoS. This may require deriving the relationship between maneuvering time and fuel consumption through simulation of HDPL scenarios. In fact, in order to calculate the time required to actually achieve the fuel savings, the maneuvering must be taken into account. The closing and opening maneuvering times must be derived from the future favorable QoS time series. In addition, if the maneuver consumes fuel, the embodiments may also take its compensation into account by deducting the compensation time. The sum of the maneuvering time and the compensation time thus drives the PQoS prediction range. Another important feature of this approach is the development of control strategies for the two maneuvers involved.

[0116] To derive the predicted range requirements for AQoSA, a standardized platooning scenario on a closed loop is considered. This allows running the same scenario independently of the actual maneuvering time, assuming that the platoon will achieve the formation change within the given time. The scenario is divided into five phases: (i) platooning; (ii) IVD reduction; (iii) HDPL; (iv) IVD increase; and (v) platooning. In the first phase (i), the platoon targets an IVD associated with a normal platooning use case, such as 30m. Then, it is assumed that the platoon receives information about a favorable QoS in the future and triggers a reduction in IVD (ii) in order to reduce the drag experienced by the follower vehicles, with the goal of reducing overall fuel consumption. In the HDPL phase (iii), the platoon targets an IVD of 5m. The platoon then realizes the degradation of QoS and triggers an increase in IVD (iv). When the maneuver is achieved, that is, when the IVD returns to its original value, the platoon continues driving (v). In an embodiment, the platoon consists of five trucks that are driven on a test loop, as illustrated in Figure 4. FIG. 4 illustrates scene snapshots: (a) a simulation model of the test track facility, and (b) a zoom-in on a platoon of five trucks in this embodiment.

[0117] The scenario is divided into three periods of 100 s: (i); (ii-iii) and (iv-v). The actual duration of phases (ii) to (iv) depends on the implemented maneuvering strategy and is the result of these experiments. In addition to achieving a standardized experimental process, phases (i) and (iii) also provide stabilization periods. The data collected in this time-triggered implementation allows the development of strategies to handle actual PQoS time series. Each experiment actually generates two maneuvering times and two maneuvering fuel consumptions.

[0118] In this study, the longitudinal controller for IVD adaptation proposed in I. Navarro, F. Zimmermann, M. Vasic, and A. Martinoli, "Distributed graph-based control of convoys of heterogeneous vehicles using curvilinear road coordinates" (19th IEEE Int. Conf. Intell. Transp. Sys. (ITSC), 2016, pp. 879–886) is extended in terms of maneuver time and fuel efficiency. The advantage of this control strategy compared to conventional cooperative adaptive cruise control (CACC) is that the platoon is more robust to unstable communication links. The first study of the requirements for a communication system using such a control strategy for emergency braking was presented in “Simulation of cooperative automated driving by bidirectional coupling of vehicle and network simulators” by I. Llatser, G. Jornod, A. Festag, D. Mansolino, I. Navarro, and A. Martinoli (IEEE Intell. Veh. Symp. (IV), 2017, pp. 1881–1886).

[0119] In an embodiment, the distributed control algorithm is based on the Laplace control principle, see M. Mesbahi and M. Egerstedt, "Graph theoretic methods in multiagent networks" (Princeton University Press, 2010, vol. 33).

[0120] Each vehicle calculates its target speed using the following equation:

[0121]

[0122] in s is the position vector in the Frenet-Serret frame, b is the bias vector that defines the formation, and v g is a vector of scalar target group velocities. L is the Laplacian matrix, which is defined as:

[0123]

[0124] where I and W are the incidence matrix and weight matrix of a graph consisting of a vehicular ad hoc network (VANET), respectively. The nodes of the graph are vehicles and its edges are communication links.

[0125] Embodiments may enhance the offset and bias mechanisms. In the original algorithm proposed in "Distributed graph-based control of convoys of heterogeneous vehicles using curvilinear road coordinates" by I. Navarro, F. Zimmermann, M. Vasic, and A. Martinoli (19th IEEE Int. Conf. Intell. Transp. Sys. (ITSC), 2016, pp. 879–886), the offset and bias are calculated relative to a reference vehicle that is the front vehicle. This definition applies to the creation and maintenance of a convoy, even more so when the convoy consists of a variety of vehicles. When it comes to the modification of the IVD, it is interesting to optimize the position of the reference point according to the maneuver performed (increasing or decreasing the IVD), the aerodynamic parameters, and the optimization goal. In fact, on the one hand, having the reference in the middle of the row divides the maneuver time by two. On the other hand, it is more fuel efficient to have the reference at the back when reducing the IVD and at the front when increasing the IVD. FIG. 5 illustrates a simple example of a combination of goals and maneuvers.

[0126] FIG5 illustrates reference placement in an embodiment for: (a) optimization of fuel consumption with the reference at the rear, (b) optimization of maneuver duration with the reference at the center, compared to (c) a conventional reference at the front, during an IVD closing maneuver. FIG5 includes, for each reference example, a fuel consumption time series, a speed time series, and a diagram of the effect of the reference on the maneuver. In the fuel consumption time series, a simple curve represents the instantaneous consumption without considering air resistance. In the upper right corner, the overall consumption of the row over a 50 s experiment is also given. The instantaneous fuel consumption taking into account the reduced air resistance caused by the preceding vehicle is represented by a low opacity curve. In the speed time series, the maneuver time T is indicated. M In the reference illustrations (the third one for each subsection), the driving direction is from left to right and the horizontal arrow indicates the direction of acceleration.

[0127] In the following, variable parameters for evaluating the relationship between fuel consumption and maneuvering time in an embodiment are described. In an embodiment, calculation methods for these two objectives are then described before presenting a fuel consumption model evaluation for highlighting the benefits of HDPL.

[0128] In some embodiments, the control strategy has three independent features: (i) row controller, (ii) local controller and (iii) command input validation. The row controller will be described later, which enables group behavior. The local controller converts the velocity received from the row controller into acceleration commands. Finally, the command input validation ensures that the velocity and acceleration commands do not violate the dynamic constraints.

[0129] Figure 6 The platoon input parameters and time targets in the embodiment are shown. The maneuvers corresponding to these speed profiles use the center truck as the reference, so r = 0.5. The control gain k is depicted for trucks 0 and 4. l The velocity and acceleration are limited to v g ±Δv and ±a max Although the maneuvering time is defined by the deviation threshold, it can also be Figure 6 Maneuvering time was observed in .

[0130] Figure 6 The speed curve using IVD changes illustrates the parameters described below. In an embodiment that allows a compromise between these two goals, different platooning strategies can be applied. In a platoon with an odd number of vehicles, three example strategies are front, middle, and rear reference, which are further referred to as This variable is used as To find the reference vehicle index, where N v Indicates the number of vehicles in the row (the numbering starts at 0). In general, the reference point is not limited to vehicles and can be placed anywhere within the row.

[0131] The control gain is a parameter of the local controller and determines how tolerant the algorithm is to differences between the current formation and the target formation. Within a reasonable range of values, larger control gains k l Faster maneuvers will be achieved at the expense of potentially greater fuel consumption.

[0132] The vehicles in the platoon have inherent dynamic limitations, such as maximum speed and acceleration capabilities, due to their mechanical properties. In addition to these, some maximum and minimum speeds around the target platoon speed, as well as maximum and minimum accelerations are imposed. These two sets of parameters determine the shape of the speed profile during the maneuver, in terms of the height of the plateau and the slope of the speed increase and decrease. To reduce the complexity of the problem, at the target speed v g =25m / s Use symmetrical value around v g ±Δv, and at 0m / s 2 Use symmetric value ±a around max These two parameters have different effects on the objectives. For example, a sharp acceleration has a relatively high cost in terms of fuel consumption, but reduces the speed change time. In addition, a smaller speed operating area limits the consumption of the acceleration phase, but produces a longer time for the maneuver.

[0133] Some embodiments may target two objectives, minimizing maneuvering time and minimizing fuel consumption. The former is of interest to communication systems because it drives the prediction range of PQoS. The latter is an inherent goal of platoon systems, namely reducing fuel consumption.

[0134] 1) Maneuver duration: Maneuver time T m Can be defined as triggering IVD changes (T t ) and the time it takes to reach that goal. This occurs when:

[0135] ,

[0136] For the local graph of node i, (see equation (1)), and ε is the tolerance value, which is 1m in the presented results. is related to the average deviation between the IVD and the target IVD for each vehicle. The duration is derived as Figure 7 As shown in the figure. Figure 7 FIG. 4 illustrates the graph deviation as a function of time and the maneuver duration T in an embodiment. m For presentation purposes, the values ​​have been exaggerated. ε The value of .

[0137] The observed fluctuations (e.g., around 12 s for trucks 0 and 3) correspond to changes in the number of nodes in the local graph, usually an increase during the IVD reduction maneuver. Figure 6 The maneuver duration is also indicated in , since the stabilization of the truck speed also corresponds to the achievement of the target formation.

[0138] 2) Maneuver Cost: The maneuver cost C is the total fuel consumption of the platoon during the maneuver, expressed as the difference from the fuel consumption of the same platoon that did not perform the maneuver. Therefore, each set of parameters is applied a second time to obtain a baseline. Thus, in some embodiments, the relative fuel consumption is calculated as follows:

[0139]

[0140] Where C is the relative maneuver cost, C k is the total fuel consumption during the maneuver, where C m and C b Corresponding to maneuver and baseline, respectively.

[0141] Similarly, C i,k and c i,k It's a truck i To calculate the instantaneous consumption of the truck, some embodiments rely on the fuel consumption model of the traffic simulator SUMO, namely HBEFA3 / HDV_TT described in "COLOMBO Deliverable 4.2: Extended Simulation ToolPHEM coupled to SUMO with User Guide", S. Hausberger and D. Krajzewicz, February 2014.

[0142] It is then corrected for air drag gains by using the recorded time series for fuel consumption, speed, acceleration and IVD and a drag model.

[0143] In the following, the fuel consumption model used for HDPL in some embodiments will be described in more detail. The expected benefit of truck platooning is the reduced overall fuel consumption. It is achieved by the reduced air drag experienced by each platoon member. Fuel consumption models integrated in microscopic vehicle simulators typically take into account individual vehicle parameters such as speed and acceleration. In HDPL systems, the air drag experienced by a truck also strongly depends on the other platoon members. Therefore, an air drag correction strategy can be developed that takes the following as input for each truck:

[0144] (i) Fuel consumption taking into account individual air resistance;

[0145] (ii) IVDs; and

[0146] (iii) position within the row.

[0147] First, the model proposed by Alam et al. [1],

[17] is linearized, [1] A. Al Alam, A. Gattami, and KH Johansson, "An experimental study on the fuel reduction potential of heavy duty vehicle platooning" (13th Int. IEEE Conf. on Intell. Transp. Syst. (ITSC), 2010. IEEE, 2010, pp. 306–311),

[17] A. El Alam, "Fuel-efficient heavy-duty vehicle platooning" (Ph.D. dissertation, KTH Royal Institute of Technology, 2014), highlighting the relationship with the HDPL system parameters. Second, it is explained how the input time series is processed to take into account the reduced air drag.

[0148] Finally, the corrected fuel consumption is obtained which can be used in equation (3) above. Alam's model contains environmental as well as truck-specific parameters:

[0149] = F 牵引 - F 滚动 - F 重力 - F 空气阻力 - F 制动 , (6)

[0150] The acceleration is a And the vehicle mass is m Note that this assumes that active traction and braking forces cannot be experienced simultaneously. Rolling resistance, the effects of gravity, and air resistance are given by the following equations:

[0151] F 滚动 =

[0152] F 重力 =

[0153] F 空气阻力 =

[0154] in, c r , g and They are the rolling resistance coefficient, gravity constant, and road slope. In addition, c D is the air resistance coefficient, A a is the maximum cross-sectional area of ​​the vehicle, and is the air density. Note that because In our simulation setup, it is empty, so F 重力 is also empty. For the sake of generality, it will continue to be expressed in the following. Nonlinear air resistance ratio It is the essence of fuel saving. is a function of IVD, but also takes into account the position of the vehicle within the platoon. When no resistance is experienced, →0; when the vehicle cannot benefit from reduced air resistance due to the preceding truck, = 1. Its derivation is presented in P. Vegendla, T. Sofu, R. Saha, MM Kumar, and L.-K. Hwang, “Investigation of aerodynamic influence on truck platooning” (SAE Technical Paper, Tech. Rep., 2015).

[0155] Therefore, the fuel consumption corrected for instantaneous air resistance is c i,k can be expressed as:

[0156]

[0157] It only considers the instantaneous fuel consumption of individual vehicles, that is, under constant air resistance =1, is the savings from IVD-related air resistance:

[0158] ma + F 滚动 + F 重力 + F 空气阻力

[0159]

[0160] In (12), is obtained by linking IVD, position within the row, and a table of speed and air drag ratios provided in "Aerodynamics of road vehicles" by H. Wolf-Heinrich and SR Ahmed (Society of Automotive Engineers, 1998) and by Figure 8 As shown in the picture. Figure 8 The air drag ratio as a function of IVD for five trucks in a platoon is illustrated in an embodiment. It is assumed that the fourth and fifth trucks experience the same air drag ratio as the third truck. An air drag ratio of 1 corresponds to a truck driving alone, which does not experience any benefit from the other trucks in the platoon.

[0161] It should be noted that due to ,therefore . This is illustrated in Figure 5 using flat and transparent curves, respectively. and c i,k . is the vertical distance between the corresponding curves. By using the relative cost function introduced in equation (5), the gain of HDPL can be studied and different maneuvers can be compared. Using this model, the net gain of performing HDPL in the case of five trucks can be calculated based on the initial and final distances, and Fig. 9 The results were reported. Fig. 9 Relative fuel savings as a function of final IVD for various initial IVDs in an embodiment are shown. The approximate correspondence in percentage is given as the secondary axis.

[0162] In particular, in the case of 5 trucks, the HDPL is 2.2 mL / s at 5 m compared to 30 m in platoon. These values ​​will be used to calculate the compensation time for the maneuvering fuel investment in the following scenarios. This then allows the fuel investment to be converted into compensation time and vice versa. This is particularly useful when it comes to optimization for two objectives - minimizing fuel consumption and maneuvering time.

[0163] First, the simulation parameters will be described. Two sets of experiments will be performed. The first set allows the relationship between maneuver duration and fuel consumption to be closely studied for one combination of initial and final IVD. The effect of fuel consumption on this relationship is also studied. Using the insights gathered from these results, the relationship can be generalized for multiple combinations of initial and final IVD.

[0164] HDPL scenarios can be run using a coupled traffic and network simulator with SUMO and ns-3. The implemented control strategy is described directly in the previous section of ns-3. Using the tracing capabilities of ns-3, instantaneous IVD, speed, fuel consumption, and map deviation time series can be collected for combinations of control parameters.

[0165] In the first phase, the impact of control parameters can be studied for specific combinations of initial and final IVDs. In the second phase, the insights gathered in the first phase can be used to extend the results to multiple combinations of initial and final IVDs. These parameters are summarized in Table 1 below, resulting in a total of 1800 experiments, including the baseline, which is as follows:

[0166] .

[0167] These time series are then processed in order to obtain the total fuel consumption and the maneuvering time. The target platoon speed was chosen to be 23m / s = 90km / h because it allows to provide a fair comparison with the actual baseline. In fact, a situation where multiple trucks are driving at a speed of 90km / h with an IVD of 30m can be observed on the highway. It has the disadvantage of creating a situation where the truck speed exceeds the maximum legal speed in some countries. Considering that autonomous driving with an IVD of 30m is not yet allowed, these results are provided for the purpose of revealing the potential of HDPL.

[0168] Some embodiments derive a strategy selection for closing and opening maneuvers from the PQoS time series. This selection is driven by two simultaneous objectives, namely the minimization (reduction) of fuel consumption and the minimization of maneuvering time. In fact, for the overall problem, the longer the maneuvering time, the higher the risk of having to perform an earlier and faster opening maneuver. The risk is higher due to the low confidence in the provided PQoS.

[0169] FIG10 shows the relationship between relative fuel consumption and maneuver time for two maneuvers when no drag correction is applied. FIG10 shows the performance evaluation without drag correction for (a) a closing gap maneuver and (b) an opening gap maneuver for different parameter combinations in an embodiment. Different labels depict different versions of the CACC algorithm: the bullet represents the traditional front reference; the square represents the center reference optimized for time; and the diamond represents the rear reference, respectively. The size of the marker is related to the maximum speed deviation. The thick curve indicates the approximate Pareto front, which is the line where an improvement in one objective is achieved as another objective deteriorates.

[0170] This set of results shows that a closing maneuver of 20s can be achieved at the expense of a 200mL increase in fuel consumption. The opposite maneuver shows similar results, with a minimum maneuver time of 30s. This larger minimum maneuver time is caused by the inherent properties of the controller, which is more flexible to IVD deviations when the IVD is larger. Another difference worth noting is that it is possible to perform an opening maneuver with negative relative fuel consumption. This may be due to the fact that some trucks slow down with low speed changes. However, maneuvers of this nature are the slowest, lasting more than 50s.

[0171] For the same Δv, the middle reference consistently outperforms the other two references, but at the expense of greater fuel consumption. As expected, “catching up” (front reference) to close the gap will consume more fuel than decelerating, and vice versa. An interesting result for the strategy selection is the approximate symmetry between the front and rear references across the two maneuvers (highlighted by the ellipses and Roman numerals in Figure 10). For example, by switching between the two references, it is possible to perform both maneuvers in 80 seconds while investing less than 300 mL.

[0172] Allowing for a larger Δ v This generally reduces maneuver time while increasing fuel consumption. This parameter has the largest and most significant impact on both objectives compared to maximum acceleration and control gain (these parameters are not highlighted in Figure 11 for readability purposes). For the second set of results, the drag correction model was applied to the time series before calculating these metrics. Figure 11 provides the relationship between the two metrics for both maneuvers.

[0173] FIG11 illustrates performance evaluation with drag correction for different parameter combinations in (a) closed gap maneuvers and (b) open gap maneuvers in an embodiment. The marker shape and size scheme is the same as the previous figure. An approximate Pareto frontier of the uncorrected results has been added with a dashed line for comparison purposes. Once the above corrections are applied, the trends highlighted in the description and analysis of the previous set of results generally remain unchanged.

[0174] When reducing IVD, a 50mL difference in the relative fuel consumption can be observed for the fastest maneuvers. The difference is smaller for slower maneuvers, which reach 140mL. This behavior is due to the fact that these maneuvers already benefit from the reduced drag when reaching their target formation in a slower way. This difference would be largely compensated if the relative fuel consumption over a fixed time period was calculated. Another noteworthy feature is that the difference for the front reference data point is larger than for the rear reference. In fact, the benefit is greater when the front vehicle approaches first.

[0175] The opening maneuvers show a smaller dependence of the correction on the maneuver time. These are stable between 50 and 60 mL for the middle and front references. Here, the rear reference maneuvers benefit the most from HDPL, since the front truck holds the small IVD longer. Figure 11 also indicates the HDPL time necessary to compensate for the fuel input (using the above value of 2.92 mL / s). Fast maneuvers tend to produce very high compensation times, prompting the need to find a compromise that takes into account the available PQoS time series.

[0176] Previous simulation results show the importance of HDPL maneuver reference and speed bounds in terms of duration and fuel consumption in opening and closing maneuvers. An obvious limitation of these results is the fact that decisions about the benefits of HDPL can only be made based on one initial IVD and one final IVD. The initial and final IVDs can be studied separately. d i and d f New combination: {10, 15, 20, 25, 30}m and {5, 10, 15, 20, 25}m, where d i > d f , as reported in the table above along with other Phase II parameters. Selected results of this new set of experiments are presented in FIG. 12 .

[0177] FIG. 12 shows selected results of extended embodiment simulations: for fixed initial IVD for (a) anterior reference, (b) intermediate reference, and (c) posterior reference d i = 30 m of closing maneuver; and (d) for a fixed final distance d f = 5 m intermediate reference opening maneuver. Marker shapes correspond to final IVD (ac) and initial IVD (d). Marker size and maximum velocity deviation Proportional to m / s.

[0178] Sub-figures (ac) show the d i = 30 m for a closed maneuver and the relationship between the maneuver time. A general trend in the dependence between the two metrics can be observed: the metric pairs are along each pair { d i , d f}. The trend is similar to the approximate Pareto front described in the previous result set. It can also be observed that the parameter Δv drives the value along the front, i.e., increasing Δv reduces the maneuver time while increasing the relative fuel consumption. These three sets of results also illustrate the main effect of the reference: the rear reference reduces fuel consumption compared to the front reference while having little effect on the maneuver duration, while the middle reference slightly reduces fuel consumption but divides the maneuver time almost by two.

[0179] Sub-figure (d) provides the d f =30m for the intermediate reference opening maneuver. Although less pronounced, the above trend also holds. These observations can be used as motivation for developing a model that will d i , d f}Yes, reference r The relative fuel consumption and maneuver duration are provided in the case of the maximum speed deviation Δv.

[0180] Given a PQoS time series, an embodiment can provide an optimal or improved strategy for HDPL, taking into account the fuel consumption and duration of closing and opening maneuvers. The first step is to convert the PQoS time series into the minimum allowed IVD that takes into account the safety of the platoon. An example of this conversion is provided in "Predictive Quality of Service: Adaptation of Inter Vehicle Distanceto Packet Inter-reception Time for HDPL" by A. Pfadler, G. Jornod, A. El Assaad, and P. Jung (IEEE 91st Veh. Technol. Conf. (VTCSpring). IEEE, 2020, under review) for the inter-packet reception time (PIR) QoS indicator.

[0181] The result of this transformation step can be represented as a pair of values: T F and the minimum IVD allowed d m . T F Indicates that allowed d m The purpose is to select the final IVD for closing and opening maneuvers. d f And the corresponding r and Δ vIf possible during the duration T F to perform the opening and closing maneuvers within a certain time frame and save fuel, the platoon should only change its IVD to d f ,in d i > d f ≥ IVD min . d i represents the initial IVD and is the IVD of the platoon before it begins the maneuver. It is assumed that after the HDPL maneuver, the platoon targets the same distance.

[0182] One of the main benefits from HDPL is fuel savings. Embodiments may aim to maximize this benefit while taking into account the constraints imposed by the QoS time series, i.e. d m and T F , which are the minimum drivable IVD and the time during which it is possible to drive with the minimum drivable IVD. The optimization problem can be formulated as:

[0183]

[0184] in F S It is the effective fuel savings after deducting all maneuvering inputs. F S It can be derived as a function of different maneuvering parameters. A continuous model for the relationship between maneuvering fuel input and time can be proposed. An evolutionary algorithm can be applied to solve the problem and its results are presented.

[0185] In order to provide a solution to (13), we need to study F S Dependence on maneuver parameters. F S Is the selected target IVD d f Fuel saving rate And effective fuel saving time T H Function of

[0186]

[0187] in ,like Fig. 9 Just as described. Fuel efficient and saves time T His equal to the time during which the HDPL benefits from a small IVD. It takes into account all maneuvering investments, such as the cost of closing and opening the gap. These maneuvering investments are considered as compensation times for the opening and closing maneuvers. The compensation time corresponds to the time during which the HDPL must maintain its target IVD to save enough fuel to compensate for the maneuvering fuel investment. The compensation times for the closing and opening maneuvers can be expressed as and The two are calculated by comparing the relative fuel consumption of maneuvers. C Divide by fuel savings The fuel saving rate is Depends on the selected d f .

[0188] .

[0189] Taking into account maneuvering time, T H Can be expressed as favorable time T F Function:

[0190]

[0191] in and are the durations of the closing and opening maneuvers, respectively. The results presented in the previous section support that for certain combinations of parameters, the compensation time can be null or negative. It is in fact possible not to consume any fuel or even to save fuel during the maneuver, in which case T H The following table summarizes the different durations involved and Fig.13 Their dependence on IVD is illustrated.

[0192] .

[0193] Fig.13 is an illustrative representation of IVD during transitions between platoon to HDPL and back to platoon from HDPL in an embodiment. It shows IVD during closing and opening maneuvers, compensation time for each maneuver, and fuel savings duration. To calculate relative fuel consumption, a platoon that does not perform any maneuvers can be used as a baseline. When the reference is the same for opening and closing maneuvers, the overall travel distance remains the same as the baseline. However, when a different reference is used, the travel distance is different (see Figure 3 and 14 ), and needs to be taken into account. In fact, the platoon will consume fuel to travel a positive distance difference, or save fuel when the distance difference is negative.

[0194] Fig.14 A diagram showing the effect of a reference combination on the final distance budget in an embodiment. The example shows a platoon of three trucks performing an opening maneuver and a closing maneuver with different references. Depending on the combination, the platoon will have a "win" or "lose" distance compared to the baseline, just as described above with respect to Figure 3 To take this into account in the analysis, the travel distance compensation time is introduced . which corresponds to the time during which the HDPL must maintain its target IVD to compensate for the fuel associated with the range difference. Has the opposite sign to the distance difference. Negative is equivalent to expanding T H . is calculated as follows:

[0195]

[0196] in It's a truck i In IVD d = d f as well as r c , r o = {0, 0.5, 1} for the baseline instantaneous fuel consumption.

[0197] From the results in the previous section, it can be seen that the maneuvering time and maneuvering fuel consumption depend on the reference, maximum speed deviation, initial and final IVD. Compared with these two parameters, the gain and maximum acceleration deviation have a negligible effect and are ignored in the following. The compensation time depends on the fuel consumption and the fuel saving rate, which itself depends on the initial and final IVD. The optimization problem can be formulated in the following form:

[0198]

[0199] where the indices c and o correspond to closure and maneuver. d i The original target IVD fixed to the row was obtained In order to optimize the fuel saving given the PQoS time series, the parameters need to be related to the duration of the maneuver and the fuel consumption. In an embodiment, the fuel consumption can be taken into account as the compensation time. , , and Continuous model.

[0200] In the following, maneuvering performance modeling will be described for an embodiment. In describing the extended results of the simulation, and in the previous paragraphs, the design of a model that generates relative fuel consumption and maneuvering duration was prompted. The modeling should be based on initial and final IVD, reference and maximum speed deviations.

[0201] In some embodiments, a least absolute shrinkage and selection operator (Lasso) model of R. Tibshirani, "Regression shrinkage and selection via the lasso" (Journal of the Royal Statistical Society (Series B), vol. 58, pp. 267-288, 1996) is fitted to the data. Lasso is a regression method that performs variable selection and regularization steps to improve prediction accuracy. On top of least squares regression, Lasso constrains the sum of the absolute values ​​of coefficients below a threshold. As a result, coefficients of parameters that have little or no effect on the predicted variable are set to zero or close to zero. This constraint implements both regularization and variable selection. Consider p Covariates , and a single result y , the Lasso parameters are calculated using the following minimization :

[0202]

[0203] in β is the coefficient vector, X is the observation matrix, and λ is a tuning parameter, i.e. a positive or null constant representing the contraction penalty. Lasso is a univariate analysis method: therefore, the two observed variables - duration and relative fuel consumption must be modeled separately.

[0204] When executing a closing maneuver, the platoon saves more fuel at the end of the maneuver than at the beginning. This is reinforced by the shape of the IVD time series (which can be seen from Figure 7 12 (b) and (d)). In fact, the longer duration of the maneuver occurs in the case of small IVDs due to the behavior of the proportional part of the controller. The opening maneuver is the opposite, since it starts with the most favorable IVD and then moves quickly away from it. Under the same conditions (initial and final IVD, and reference), both maneuvers take approximately the same duration, but the opening maneuver is more economical than the closing maneuver, as it is possible to notice when comparing Figures 12 (b) and 12 (d).

[0205] Each maneuver reference induces a different behavior in terms of vehicle movement. This affects the aerodynamic conditions and therefore the fuel consumption on the one hand, and the maneuver duration on the other hand. The front and rear references generally produce similar durations and different consumptions (which has a symmetry, as highlighted in Figure 10). The middle reference generally produces the shortest maneuver, as shown in Figure 11.

[0206] Reference (anterior, middle, posterior) and maneuver type (open and closed) can be considered categorical variables. Accounting for these features as continuous variables in the Lasso model would greatly impair the performance of the modeling. These categorical variables can be considered by fitting six models (three references times two maneuvers) for two target values. This results in 12 regression models.

[0207] The inputs to these regressions are the initial and final IVDs (respectively d i and d f ), and the velocity boundary Δ v . Observing the shape of the different points in the results presented in the previous section, the target variable shows a polynomial dependency on the velocity bounds. Given the regularization properties of the Lasso regression, it is possible to employ a polynomial combination of our features without the risk of overfitting. In order to establish the degree of the polynomial combination and the shrinkage penalty λ , an exhaustive grid search can be performed using k-fold cross validation. The coefficient of determination r 2 can be used as a scoring function, and k = 3. If the degree of the polynomial is almost always 3 for the 12 models, then λ The best value of varies. This reinforces the use of a grid search over all 12 models, rather than fitting hyperparameters on just one model.

[0208] For each combination of categorical variable, reference and maneuver type, a continuous model linking the target feature with the three predictors can be obtained. Figure 15 presents exemplary results for an anterior reference closing maneuver and a mid-reference opening maneuver.

[0209] FIG. 15 illustrates a surface model created by combining the graphs of duration and relative fuel consumption of a maneuver in an embodiment using a Lasso regression model. The solid line represents the d i and d fResults of the combined simulated models. The dashed lines show some examples of maneuvers that were not simulated and the performance obtained using the continuum model. The top (solid) or bottom (dashed) of each corresponding line shows the fixed d i The solid line in (a) corresponds to the model for the same d i and d f The continuous result of the combination, and the solid line in (b) corresponds to the model for the same d i and d f Combined continuous results.

[0210] In this result set, d f = 5m is fixed for presentation purposes. To illustrate the interpolation capability of the resulting model, the results of the model for unknown values ​​are shown in addition to the results for the simulated initial IVD values. The evolution of the position of the curve depending on the initial IVD confirms the insights gathered in the analysis of the results presented in Figure 12.

[0211] The obtained Lasso regression allows to obtain the relative fuel consumption and maneuver duration for each combination of maneuver (closed and open), reference (front, middle, rear), and initial and final distances, where , ,and d i > d f Applying equation (15) to the obtained fuel consumption, we can get Any combination of F S .

[0212] To verify the latter constraint, the obtained regression model can be used to calculate the fuel consumption and maneuver duration. For functional safety reasons, it can be assumed that the closing maneuver should not be started before the communication requirements are met. Similarly, the opening maneuver should be completed while the communication requirements are met.

[0213] Given a PQoS time series and converting it to a minimum IVD time series, the time series can be described by two values, d m and T F. The objective function described in the previous paragraph has the disadvantage that it is not differentiable when considering different references. This property of the objective function prevents the use of any gradient-based methods, such as gradient descent and quasi-Newton methods. In order to find the five parameters that maximize fuel savings in (18), some embodiments use an intensive search optimization algorithm.

[0214] Particle swarm optimization (PSO), an evolutionary algorithm that performs iterative stochastic optimization, can be used; see J. Kennedy and R. Eberhart, “Particle swarm optimization” (1995 Int. Conf. Neural Networks (ICNN'95), vol. 4, November 1995, pp. 1942–1948), and Y. Shi and R. Eberhart, “A modified particle swarm optimizer” (1998 IEEE International Conference on Evolutionary Computation Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98TH8360), May 1998, pp. 69–73).

[0215] In PSO, the set of candidate solutions is called a particle swarm. Particles navigate in a multidimensional search space according to their positions and velocities. At each step, these particles are updated depending on their own best experience, the best experience of the population, and random parameters. The combination of global and local optimality and the use of randomness in the update attempt to avoid the search from falling into a local minimum. However, PSO is a meta-heuristic algorithm: it can obtain good solutions, but it cannot guarantee a global optimal solution.

[0216] In each iteration k When , the particle position vector x and velocity vector v are updated according to the following equations:

[0217]

[0218] in w is the weight factor and corresponds to the inertia of the particle; c 1 and c 2 are constants and determine the effects of particle and group experience separately; r 1 and r 2are two random numbers drawn from a uniform distribution each time equation (21) is applied. U (0,1); p lb (Correspondingly, p gb ) is the position where the particle (respectively, the particle population) has the best performance. w , c 1 and c 2 is a hyperparameter that needs to be set in advance. In addition, PSO also needs to define the population size N and the number of iterations I . Search space dimension D is inherent to this optimization problem. Finally, in Equations (20) and (21), the position is constrained to the bounding box and each velocity component is constrained to the absolute velocity maximum. Algorithm 1 summarizes the operation of the algorithm, with position, velocity, and optimal position updates.

[0219] Algorithm 1: Particle Swarm Optimization (PSO)

[0220] 1 objective function g(x);

[0221] enter: N , I , D , v max

[0222] Output: Global best position p gb

[0223] 2 Initialize position and velocity;

[0224] 3 for each iteration k = 1 … I do

[0225] 4 for each particle i = 1 … N do

[0226] 5 Calculation of objective function ;

[0227] 6. Achieve global and local optimal positions;

[0228] 7 for each dimension d = 1 … D do

[0229] 8 Calculation speed

[0230] 9 Update particle position

[0231] 10 end

[0232] 11 end

[0233] 12 end

[0234] A grid search based approach for hyperparameter optimization can be used. The objective function can be set as a scoring method. The search grid can be applied to the favorable duration T F , Initial IVD d i The selected hyperparameter tuple is the one that makes PSO produce the best solution most frequently.

[0235] In some embodiments, the ultimate goal may be to select a set of parameters for closing and opening maneuvers that will maximize fuel savings and thereby derive a predicted range value. To this end, different sets of T F and d m to perform optimization. It can be used as a d m Function T F The fuel savings are observed and are in the order of magnitude of the predicted range.

[0236] Fig.16 shows the minimum IVD d m =5m and initial IVD d i =30m fuel saving optimization results. Fig.16 The embodiment for d f = 5m. Each curve represents a combination of references for closing and opening maneuvers. It can be observed that the combination of the middle references gives the highest savings at most of the time. Just below it, at most of the time, is also the combination of the middle and rear references. Finally, the traditional front reference performs the worst.

[0237] Here, 5m is the minimum constraint, and the row can d m and d iThe maximum fuel savings can usually be achieved by a combination that includes the proposed intermediate reference strategy. T F =100s, the fuel savings are very close to 0%, providing approximate minimums for the favorable time and for the prediction range. In general, all strategies outperform the traditional method using a front reference for all maneuvers. Finally, all curves tend to a plateau. This plateau represents the maximum fuel savings percentage, which is reached when the maneuvering cost becomes negligible compared to the fuel saved at low IVD. This occurs around 600s, which means that at a speed of 90km / h, the platoon needs to travel 15km before the maneuvering cost becomes negligible.

[0238] Embodiments are based on the above-described investigation of maneuvering strategies to achieve fuel efficiency when performing high density platooning. Embodiments may use the concept of reference placement in a platoon control system to balance the two objectives of this problem - fuel consumption and minimization of maneuver duration. The relationship between these two metrics may be studied as a function of different maneuvering parameters. Embodiments may consider not only the fuel savings achieved by driving with a smaller head spacing, but also the fuel savings achieved during maneuvering by developing a fuel consumption correction method. It can be shown that some strategies achieve fuel savings when performing the maneuver, but at the expense of a longer maneuver duration. The correction model also allows the fuel consumption of the maneuver to be linked to the compensation time.

[0239] In some embodiments, using these results, the required prediction horizon for a predictive quality of service system can be calculated. This time depends on how far in the future a favorable quality of service will occur, as it drives the minimum maneuvering time. In some embodiments, an application derived from this prediction horizon on an inter-packet reception time prediction time series can be used. It can be shown that, considering that the quality of service is provided at no cost, in order to actually achieve fuel efficiency with a minimum IVD of 5m, the platoon needs a projection of about 100s into the future.

[0240] These results provide prediction range requirements for a quality of service prediction system within the context of high density platoon systems.Linked with a system that provides feasible inter-vehicle distances taking into account future quality of service, this decision strategy will be an enabler for agile quality of service adaptation in some embodiments.

[0241] Those skilled in the art will readily recognize that the steps of various methods described above can be performed by a programmed computer, such as the position of a time slot can be determined or calculated. In this article, some embodiments are also intended to encompass program storage devices, such as digital data storage media, which are machine or computer readable and encode machine executable or computer executable instruction programs, wherein the instructions implement some or all of the steps of the method described in this article. The program storage device can be, for example, a digital memory, a magnetic storage medium such as a disk and a tape, a hard drive, or an optically readable digital data storage medium. The embodiment is also intended to encompass a computer programmed to perform the steps of the method described in this article, or to encompass a (field) programmable logic array ((F) PLA) or a (field) programmable gate array ((F) PGA) programmed to perform the steps of the method described above.

[0242] The specification and the drawings only illustrate the principles of the present invention. It will be appreciated that those skilled in the art will be able to conceive of various arrangements that, although not explicitly described or shown herein, embody the principles of the present invention and are included in its spirit and scope. In addition, all examples recorded herein are in principle explicitly intended to be used only for teaching purposes to assist readers in understanding the principles of the present invention and the concepts contributed by (one or more) inventors to promote the art, and should be interpreted as not being limited to such specific recorded examples and conditions. In addition, all statements of the principles, aspects and embodiments of the present invention and their specific examples recorded herein are intended to cover their equivalents. When functions are provided by a processor, these functions may be provided by a single dedicated processor, by a single shared processor, or by multiple separate processors (some of which may be shared). In addition, the explicit use of the term "processor" or "controller" should not be interpreted as exclusively referring to hardware capable of executing software, and may implicitly include but is not limited to digital signal processor (DSP) hardware, network processor, application specific integrated circuit (ASIC), field programmable gate array (FPGA), read-only memory (ROM), random access memory (RAM) and non-volatile storage device for storing software. Conventional or customized other hardware may also be included. Their function may be carried out through the operation of program logic, through dedicated logic, through the interaction of program control and dedicated logic, or even manually, the particular technique being selectable by the implementer as more specifically understood from the context.

[0243] It should be appreciated by those skilled in the art that any block diagram herein represents a conceptual view of an illustrative circuit embodying the principles of the present invention. Similarly, it will be appreciated that any flow chart, flow diagram, state transition diagram, pseudo-random code, etc. represent various processes that can be substantially represented in a computer-readable medium and thus executed by a computer or processor regardless of whether such a computer or processor is explicitly shown.

[0244] In addition, the following claims are hereby incorporated into the detailed description, wherein each claim can stand alone as a separate embodiment. Although each claim can stand alone as a separate embodiment, it should be noted that although a dependent claim refers to a specific combination with one or more other claims in the claims, other embodiments may also include a combination of the dependent claim with the subject matter of each other dependent claim. Such combinations are proposed herein unless it is stated that a specific combination is not intended. In addition, it is intended that features of a claim be included in any other independent claim as well, even if the claim is not directly made dependent on the independent claim.

[0245] It is further noted that the methods disclosed in the specification or in the claims may be implemented by a device having means for carrying out each of the respective steps of these methods.

[0246] Reference numerals list

[0247] 10 Method for controlling maneuvers within a platoon of multiple vehicles

[0248] 12 Receive information related to the platoon's maneuvers

[0249] 14 Determine information about the fuel efficiency of the maneuver

[0250] 16 Decide whether to perform the maneuver based on information about fuel efficiency

[0251] 20. Device for controlling maneuvers within a platoon of multiple vehicles

[0252] 22 One or more interfaces

[0253] 24 Control Module

[0254] 100 Vehicle / Network Components

[0255] 102 Vehicles

[0256] 400 Mobile Communication System

Claims

1. A method (10) for controlling maneuvers within a platoon of a plurality of vehicles (100; 102), wherein the maneuvers include a closing maneuver for reducing a distance between vehicles, a maintaining maneuver for maintaining a distance between vehicles, and an opening maneuver for increasing a distance between vehicles, the method (10) comprising: receiving (12) information related to the maneuver of the platoon; determining (14) information about the fuel efficiency of the maneuver based on fuel consumption of maneuvering from an initial state to a maneuvering target state, maintaining the maneuvering target state, and returning from the maneuvering target state to the initial state, wherein the determining (14) includes determining an effective time of the maneuvering target state, wherein the effective time is based on the maneuvering from the initial state to the maneuvering target state, maintaining the maneuvering target state, and returning from the maneuvering target state to the initial state; and A decision is made (16) whether to perform the maneuver based on information about fuel efficiency.

2. The method (10) according to claim 1, wherein the maneuver is a high-density platoon maneuver, and for the high-density platoon maneuver, the inter-vehicle distance of the maneuver target state is determined based on the communication waiting time between the vehicles.

3. The method (10) of claim 2, further comprising determining an inter-vehicle distance for a maneuvering target state based on a predicted quality of service for inter-vehicle communications in the platoon.

4. The method (10) according to one of claims 1 to 3, wherein the information about fuel efficiency is further based on the total duration of maneuvering from an initial state to a maneuvering target state, maintaining the maneuvering target state, and returning from the maneuvering target state to the initial state.

5. The method (10) of claim 1, wherein the determining (14) comprises: The information on the fuel efficiency is determined based on the effective time of the maneuvering target state and the fuel saving rate during the maneuvering target state.

6. The method (10) of claim 5, wherein determining the effective time of the maneuvering target state comprises: Determine the actual time of maneuvering from an initial state to a target state, and the compensation time of maneuvering from the initial state to the target state, and wherein determining the effective time of maneuvering the target state includes: determining the actual time of recovering from the target state to the initial state, and the compensation time of recovering from the target state to the initial state.

7. The method (10) according to claim 6, further comprising: Different validity times of different reference points are evaluated for maneuvering from an initial state to a target state and for recovery from the target state to the initial state, wherein a reference point indicates a vehicle in the platoon that forms a maneuvering reference for other vehicles in the platoon.

8. The method (10) of claim 7, wherein different effective times result in different actual times and different compensation times, wherein at least one of the compensation times is negative.

9. The method (10) according to one of claims 1 to 3, wherein vehicles (100; 102) of the plurality of vehicles communicate with each other using a mobile communication system (400), and wherein the maneuver is determined by a predicted quality of service in the mobile communication system (400).

10. The method (10) according to claim 9, further comprising: An overall duration of the maneuver is determined based on the predicted quality of service.

11. A computer program product having a program code for carrying out the method (10) according to one of claims 1 to 10 when the program code is executed on a computer, a processor or a programmable hardware component.

12. A method for controlling a plurality of vehicles (100; 102), the motorized device (20) comprising: one or more interfaces (22) for communicating with one or more vehicles in a platoon of vehicles (100; 102); and A control module (24) configured to execute one of the methods according to claims 1 to 10.

13. A vehicle (100) comprising a device (20) according to claim 12.

14. A network component comprising the device (20) according to claim 12.

Images