EP2831861B1

EP2831861B1 - Method and system for adapting vehicular traffic flow

Info

Publication number: EP2831861B1
Application number: EP12721186.0A
Authority: EP
Inventors: Francesco ALESIANI; Roberto Baldessari
Original assignee: NEC Corp
Current assignee: NEC Corp
Priority date: 2012-03-30
Filing date: 2012-03-30
Publication date: 2022-12-07
Anticipated expiration: 2032-03-30
Also published as: WO2013143621A1; EP2831861A1; JP6183864B2; JP2015514261A

Description

The present invention relates to a method and a system for adapting vehicular traffic flow with a plurality of vehicles and at least one traffic flow regulation device outside the vehicles.
Driver assistance systems like cruise control (CC) systems and nowadays automatic cruise control (ACC) systems both increase the safety and the comfort of a driver of a vehicle. Automatic cruise control enables the driver of a vehicle to simply maintain a desired speed and additionally a desired safety distance to a preceding vehicle. The automatic cruise control determines continuously whether a vehicle is in front of the driver's vehicle, measures a distance to a preceding vehicle and adapts the speed of the driver's vehicle to maintain the desired safety distance, without the drivers need to interfere in order to avoid a collision with the preceding vehicle. Conventional adaptive cruise control systems enable further applications like vehicle platooning, safe distance automatic control and further cruise control applications.
In the United States patent US 8,010,274 B2 a vehicle driving support apparatus is described in which the present positional relationship between an own vehicle and a preceding vehicle is taken into account and following travel of the own vehicle with respect to the preceding vehicle is possible without the own vehicle coming too close to the front vehicle and smoothly. This is implemented by setting a prerequisite inter-vehicle distance to a preceding vehicle and a time to reaching this prerequisite inter-vehicle distance as a control target time. Further an estimated position of the preceding vehicle after the control target time elapses is computed as well as an acceleration from a present vehicle speed that will bring to a preset target inter-vehicle distance the inter-vehicle distance to the front vehicle at an own vehicle speed as of when the control target time elapses. This acceleration is used as a target acceleration on the basis of the present inter-vehicle distance to the front vehicle and the estimated position of the front vehicle after the control target time elapses. To follow the front vehicle an automatic braking control and automatic acceleration control are carried out.
In the non-patent literature of Nau at al., "Design and implementation of parameterized adaptive cruise control: An explicit model predictive control approach", Control Engineering Practice 18 (2010) 882-892 a parameterized automatic cruise control is described based on explicit model predictive control with a limited number of intuitive tuning variables for adaption of the automatic cruise control.
Further in the non-patent literature "Design of a Hybrid Adaptive Cruise Control Stop-&-Go system", master thesis of Roel van den Bleek, 2006 an adaptive cruise control stop-and-go application is described. A model predictive control (MPC) is used for fast sampling of real-time applications.
With the upcoming high availability of short range communication systems like wireless LAN, Bluetooth or the like the adaptive cruise control systems are more and more replaced by the so called cooperative adaptive cruise control systems (CACC) using additional inter-vehicle communication for enhancing the automatic cruise control systems.
In the non-patent literature of Alesiani F. and Franco G. "Optimal Speed Profile Trajectory Computation for Vehicle Approach at Intersections with Adaptive Traffic Control", Proceedings 16th World Congress on Intelligent Transport Systems, Stockholm, September 2009, a cooperative cruise control system is shown which enables an adaption of the speed of an vehicle inline with a traffic control strategy. A driver of a vehicle gets a speed advice taking into account vehicle position and speed, vehicle acceleration and a current and future traffic state.
In the non patent literature of Ploeg, J., Scheepers, B.T.M., van Nunen, E., van de Wouw, N. and Nijmeijer, H. "Design and Experimental Evaluation of Cooperative Adaptive Cruise Control, 14th international IEEE conference and intelligent transports systems, 2011, Washington D.C., U.S.A. a cooperative adaptive cruise control is described which addresses the problem of a limiting highway capacity of vehicles by decreasing the inter-vehicle distance while maintaining the same velocity level. Small inter-vehicle time gaps are shown to be string-stable allowing safe driving with time gaps between adjacent moving vehicles well below 1s.
The non-patent literature of Victor Gradinescu et al.: "Adaptive Traffic Lights Using Car-to-Car Communication", IEEEVTS Vehicular Technology Conference, Proceedings, IEEE, US, 1 April 2007 shows an adaptive traffic light system based on a wireless communication between vehicles and fixed controller nodes deployed in intersections. Vehicles send information on their position and speed to enable the controller to derive volume and demand measures, which are based on the number of approaching vehicles. This information, i.e. the number of approaching vehicles, is used to change the phase cycle of traffic lights. The number of approaching vehicles is computed once per cycle.
One of the disadvantages of conventional cooperative adaptive cruise control systems is, that the cooperation between different vehicles is limited. Further, the conventional methods provide in most cases only a solution with less accuracy, for example for a speed of a platoon of vehicles and in a very limited number of certain traffic situations.
It is therefore an objective of the present invention to provide a method and a system for adapting vehicular traffic flow with a plurality of vehicles and at least one traffic flow regulation device outside the vehicles, which are more flexible, can be used in a greater number of traffic situations.
It is a further objective of the present invention to provide a method and a system for adapting vehicular traffic flow with a plurality of vehicles and at least one traffic flow regulation device outside the vehicles enabling solutions of traffic-related problems with higher accuracy and higher performance, in particular energy and traffic efficiency while taking into account more traffic-related information. The invention is defined by the appended claims.
According to the invention the objectives are accomplished by a method of claim 1 and a system of claim 14.
The term "moving state" in particular in the description, preferably in the claims means current values for variables representing a current movement of the vehicle like speed, velocity or acceleration or the like. A moving state for example corresponds to a certain point in time in a moving profile.
The term "moving profile" in particular in the description, preferably in the claims means current and future moving states for example of a vehicle represented by trajectory of position and corresponding velocity and/or acceleration curves or the like.
The term "operational state" in particular in the description, preferably in the claims means current values for variables representing a current device state of for example of a traffic light signal controller. The traffic light signal controller is operated according to an operational profile.
The term "operational profile" in particular in the description, preferably in the claims means current and future device states for example of a traffic light controller represented by green, yellow and/or red phases for traffic lights controlled by the traffic light signal controller for example.
The term "moving profile impact information" particular in the description, preferably in the claims means, in particular traffic related, information which leads to a possible amendment of the moving profile and/or the moving state.
A convergence criterion may be determined according to a desired optimization problem, for example enhancing traffic throughput or the like. Iteration steps are taken in the direction of improving a solution to the optimization problem.
According to the invention it has first been recognized, that iteratively exchanging and adapting moving profiles of the vehicles and/or operational profiles of the traffic regulation devices enable a dynamic and distributed finding, calculation or determination of solutions for most of the traffic optimization problems, for example maximum throughput of vehicles at an intersection having traffic lights or the like.
According to the invention it has further been first recognized, that the method and the system are more flexible than conventional methods and systems: The present invention enables incorporating a variety of different traffic situations with greater complexity while providing a unique solution for traffic related optimization problems. Complexity means in particular complexity of traffic scenarios or traffic situations for example represented by a greater number of vehicles or a greater number of intersections with traffic lights.
Further features, advantages and preferred embodiments are described in the following subclaims.
According to a preferred embodiment an operational profile and/or moving profile impact information include signal phase information, preferably of at least one traffic light signal controller, for a vehicle. Signal phase information enable to incorporate information providing a stop & go of vehicles at intersections with traffic lights as example of a traffic infrastructure device. Of course, this is not limited to only one traffic light: A plurality of signal information for a vehicle assigned with different traffic lights en route of the vehicle to its destination may be also taken into account to adapt the speed of the vehicle, for example to use efficiently a green traffic wave provided by a progressive signal system.
According to the invention driving condition information for a vehicle include a speed profile and/or relative distances to preferably directly adjacent vehicles. This information is important vehicular related information and can therefore be used for efficiently adapting the overall vehicular traffic with a plurality of vehicles. In particular important is the relative distance to one or more preceding vehicles due to safety reasons.
According to the invention the steps a) -e) are initiated by a vehicle of the plurality of vehicles, preferably the vehicle which is the closest to the traffic flow regulation device. The vehicle, which has the shortest distance to the traffic flow regulation device, has in general also the shortest travel time to the traffic flow regulation device for example a traffic light. Therefore initiating the steps a) -e) provides a more efficient and optimized travelling of this vehicle avoiding for example a stop at traffic light. The travel time of the vehicle and if applicable other following vehicles to pass a traffic light could be minimized resulting in at least lower waiting times when the vehicle or the plurality of vehicles have to stop. Such a stop may be a result of a red signal phase of a traffic light and may be avoided or at least reduced by the initiation of the steps a)-e) in time before reaching the traffic light. This provides a more efficient usage of energy as well as it enhances the comfort of the driver(s) of the vehicle(s) like travel time, stopping time, etc. when approaching the traffic light. The speed of the vehicle(s) is then adapted according to the moving profile wherein the moving profile is adapted in such a way, that a green phase of the traffic light is reached when passing the traffic light by the vehicle.
According to the invention profile information of the current moving profile is received by the traffic flow regulation device and its operational profile is adapted according to the received information. The traffic flow regulation device, for example a traffic light controller with a corresponding assigned traffic light, may receive all speed profiles of the vehicles approaching the traffic light in a whole or in individual messages. According to the received profile(s) the traffic light controller may for example adjust or amend the green phase residual time to let the approaching vehicles pass instead of stopping the vehicles. Further the non used part of the green phase time window may be assigned to vehicles in other directions so that they may pass an intersection without further waiting by the traffic flow regulation device.
According to a further preferred embodiment the operational profile of the traffic flow regulation device is adapted according to ex-route information. Ex-route information includes for example the number of vehicles waiting for a green phase time window at other intersecting roads, the presence of vehicles in other conflicting directions or the like. This further enhances the flexibility and the efficiency of the method and system since more and more detailed information can be used for adapting the moving profiles and the operational profile.
According to a further preferred embodiment the operational profile of the traffic flow regulation device is adapted according to external physical information. External physical information may for example include energy and/or emission considerations in certain urban areas, for example particulate matter reduction and also for example the number of vehicles to pass a traffic light or the like. This enables a further enhanced flexibility of the invention by incorporating also physical information for example under environmental considerations. It is further possible to integrate vehicle counting which may be provided by other traffic measurement systems: The traffic flow regulation device then for example reduces the green phase time window based on the moving profile of the last vehicle of a vehicle platoon that will approach for example an intersection with the traffic light. The traffic flow regulation device may then choose which vehicle of the platoon will be the last vehicle allowed to pass the traffic light respectively the intersection based on the received moving profiles of the vehicles.
According to the invention an iteration of at least the steps d) - e) is again performed for the moving profiles when an operational profile was adapted. For example a traffic flow regulation device in form of a traffic light controller restarts an iterative process to obtain new speed profiles based on new green phase time windows of the traffic light. This iterative process enables an adaption like reducing, enlarging or shifting of green phase time windows, for example when an overall energy efficiency should be improved or vehicular traffic throughput should be improved. In each iteration the new proposed green phase time windows may be validated by each vehicle of the vehicle platoon. At the end of each successful negotiation iteration procedure, i.e. receiving the speed profiles, adapting the green phase time windows and transmitting the adapted green phase time windows to the vehicles, the traffic flow regulation device may confirm the accepted green phase time windows. This process may continue until at least one of the vehicles accepts the new proposed green phase time windows. Other vehicles waiting at an intersection because of a red phase time window corresponding to the green phase time windows which are not on a trajectory of the approaching vehicles may then be provided with a shortened red phase time window. The traffic flow regulation device at the intersection may also force vehicles to stop or enlarge the passing window, i.e. the green phase time windows, to let stopping vehicles to pass.
According to the invention an iteration of at least the steps d) - e) is again performed for the moving profiles and between the vehicles only. For example a vehicle of the vehicle platoon restarts the iteration in order to further optimize the overall vehicle platoon speed profile based on the speed profile information received from the other vehicles of the vehicle platoon. During the new iterations each vehicle may for example modify its speed profile based on the aggregated speed profiles of all vehicles in the vehicle platoon. One of the advantages is therefore, that for example a full iteration cycle including a traffic flow regulation device is not necessary saving calculation time and the amount of data to be exchanged. A further advantage is that the moving profiles of the vehicles could be more closer to each other, represented by smaller inter-vehicle-distances.
According to the invention a result of the iteration is transmitted to the traffic flow regulation device and the operational profile of the traffic flow regulation device is adapted according to the transmitted result. This enables a fast and easy adaption of the current device state and/or the operational profile of the traffic flow regulation device. For example a traffic light signal controller as traffic flow regulation device may use the last received speed profiles of the vehicles of a vehicle platoon to adjust the traffic light signal phases respectively their time windows. Further complicated calculations and/or transmission of information between the traffic flow regulation device and the vehicles are not necessary.
According to a further preferred embodiment in step d) constraint information and/or preference information and/or profile calculation parameters for at least one of the profiles are exchanged. The term "profile" here includes in particular moving profile and/or operational profile. Constraint information and/or preference information may include information on specific constraints for example minimum and/or maximum accelerations, preferences in trajectory information and/or speed profile calculation. Preference information may include safety enhancement, emission reduction, time efficiency, throughput maximizations for one ore more intersections, travel comfort or the like. For example a change in preferences of a vehicle may be added to a preceding or following vehicle as constraints in the opposite way. The current vehicle for example shall precede the following vehicle. Further a check may be performed whether there is an advantage for the following vehicle, for example greater part of the green phase time window left for the following vehicle. Profile calculation parameters may include emission and/or consumption parameters of the vehicle and/or a direct function specification relating emission and/or fuel consumption to current speed and/or current acceleration and/or a vehicle status, for example based on the vehicle weight representing a loaded or unloaded vehicle.
According to a further preferred embodiment the constraint information and/or the preference information for the states are weighed. This further enhances the flexibility of the invention and enables an easy adaption to a variety of different traffic scenarios, for example traffic jams in cities, in small villages or on highways.
According to a further preferred embodiment the moving profile, preferably the speed profile of a vehicle is calculated based on total acceleration, absolute speed, total travel distance, platooning of vehicles, maximum acceleration variation and/or platooning with preceding vehicles. This enables a more precise calculation of the moving profiles of vehicles, taking into account parameters being important for each vehicle as well as for traffic flow regulation devices.
According to a further preferred embodiment the moving profile, preferably the speed profile, is either calculated based on at least a partially known moving profile, preferably at least position, speed and acceleration of a preceding vehicle. This enables even a calculation of a moving profile without the need to exchange a vast amount of information for calculating the moving profile between the vehicles, therefore reducing inter-vehicle communication.
According to a further preferred embodiment a time slot is assigned to each vehicle in the traffic flow regulation device. This even enhances further the flexibility, and optimizes traffic flow, since for example between each of the vehicles of a platoon, other vehicles in other directions may receive a green light to pass the intersection. The platooning of the vehicles is only theoretically interrupted by the different time slots, however, the overall platooning of the vehicles en route to the destination is maintained.
According to a further preferred embodiment the time slot includes a safety time margin. One or more safety margins between for example the red light phase and the green light phase reduce an accident probability of the vehicles when passing the traffic flow regulation device at the intersection. Even if every switching from a red light phase to green light phase und vice versa takes longer due to the safety margins the throughput of vehicles at a traffic light with a greater travel comfort for the corresponding drivers may be enlarged.
According to a further preferred embodiment the operational profile of the traffic flow regulation device is calculated based on a previous complete iteration having fulfilled the convergence criterion and an iteration of exchanging and adapting operational profiles of adjacent traffic flow regulation devices. The traffic flow regulation device is then enabled to take into account both the previous iteration of speed profile exchanged among the vehicles of a platoon and an iterative exchange of traffic light phases among adjacent intersections each having traffic flow regulation devices. This provides a further enhancement for providing for example traffic throughput at intersections or enabling a large green traffic wave.
According to a further preferred embodiment of the system of claim 14 the moving profile calculations means are provided in three modules, wherein the first module is operable to perform the steps a) to e) of claim 1 for a vehicle being adjacent to the traffic flow regulation device, the second module is operable to perform the steps a) to e) for a vehicle having a preceding vehicle in front, and the third module is operable to combine results from the first and second module. For example the first module may compute an optimal speed in the scenario, where there are no vehicles in front. The second module computes speed and acceleration based either from ranging measurement systems like a onboard vehicle radar system or the like or from direct communication with preceding vehicles keeping platooning and safety distance. The third module then integrates both information of optimal speed and acceleration based on the distance information to the preceding vehicle.
According to a further preferred embodiment the first module is operable to perform module predictive control and the second and third module are operable to perform fuzzy logic control. Fuzzy logic for example may determine solutions for problems with reasoning that is approximate rather than exact.
There are several ways how to design and further develop the teaching of the present invention in an advantageous way. To this end it is to be referred to the patent claims subordinate to patent claim 1 and 14 on the one hand and to the following explanation of preferred examples of embodiments of the invention, illustrated by the drawing on the other hand. In connection with the explanation of the preferred embodiments of the invention by the aid of the drawing, generally preferred embodiments and further developments of the teaching will we explained. In the drawings

Fig. 1: is illustrating vehicles approaching a signaled intersection;
Fig. 2: is illustrating vehicles with vehicle related input parameters;
Fig. 3: is illustrating a schematic view of a speed profile computation according to a first embodiment of the present invention;
Fig. 4: is illustrating a speed profile with green phases of a traffic light;
Fig. 5: is illustrating a vehicle approaching successively different traffic lights;
Fig. 6: is illustrating a method according to a second embodiment of the present invention;
Fig. 7: is illustrating signal phase adjustments based on speed profiles;
Fig. 8: is illustrating a method according to a third embodiment of the present invention;
Fig. 9: is illustrating a fourth embodiment of the present invention;
Fig. 10: is illustrating an effect of uncertainties in a vehicles trajectory;
Fig. 11: is illustrating effects of different weight choice;
Fig. 12: is illustrating a signal phase split;
Fig. 13: is illustrating signal phase start and end time with and without safety time margins;
Fig. 14: is illustrating a partial common driving path of two vehicles;
Fig. 15: is illustrating a system according to a fifth embodiment of the present invention;
Fig. 16: is illustrating a flow chart according to a sixth embodiment of the present invention; and
Fig. 17: is illustrating a flow chart according to a seventh embodiment of the present invention.

Fig. 1 is illustrating vehicles approaching a signaled intersection.
In Fig. 1 are shown three vehicles V1, V2, V3 moving on a road R from left to right. The first vehicle V1 has a distance D12 to the second vehicle V2 which precedes the first vehicle V1. The third vehicle V3 precedes the second vehicle V2 and has a distance D23 to the second vehicle V2. The third vehicle V3 is closest to the traffic light TL and has a distance DVTL to the traffic light TL.
In Fig 2 vehicle-related input parameters IP are assigned to each of the vehicles V1, V2, V3, wherein

s _{k 0}[t ₀], v _{k 0}[t ₀], a_k0 [t ₀] is the current position, speed and acceleration of the vehicle
t=t₀+1,..., t₀+T is the time at future points in time

In Fig. 3 a speed profile computation is performed with input parameters IP

$s_{j}^{G}, t_{j}^{G}$
, j = 1,...,M is the set of green signal phases, position and time, where M is the number of approaching intersections
s _{k 0}[t ₀], v _{k 0}[t ₀] is the current position and speed of the vehicle
s_k [t],v_k [t],t≥t ₀ ,k<k ₀ the current and future position of the preceding vehicles
d_k [t],Δv_k [t],t≤t ₀ ,k=k-1 is the current and past distance and speed difference to predecessor vehicle
t=, to, t₀+1,..., t₀+T is the time at current and future points in time

s _{k 0}[t],v _{k 0}[t], a _{k 0}[t],t ≥ t ₀ is the future position, speed and acceleration of the current vehicle

The output parameters OP therefore denote the future position, speed and acceleration of each of the vehicles V1, V2, V3.
A speed profile computation with the input parameters IP and the output parameters OP may be implemented in the following way: $\begin{array}{l} \min \{λ\} \{_{1} {‖ a^{k} ‖}_{+}^{2} + λ_{2} {‖ v^{k} ‖}^{2} - λ_{3} s^{k} [T] + λ_{4, i} ({‖ v^{k} [t_{i, 1}^{G}] - 1 ν_{opt} ‖}^{2} + {‖ s^{k} [t_{i, 1}^{G}] - s_{i}^{G} ‖}^{2}) + λ_{5} {‖ {Da}^{k} ‖}^{2} + λ_{6} {‖ e^{k'} ‖}^{2}\} \\ s . t . {\begin{cases} 0 \leq v_{MIN}^{k} \leq v^{k} \leq v_{MAX}^{k} \\ a_{MIN}^{k} \leq a^{k} \leq a_{MAX}^{k} \\ v^{k} = 1 ν_{0}^{k} + {La}^{k} \\ s^{k} = 1 s_{0}^{k} + l ν_{0}^{k} + {Ka}^{k} \\ s^{k} [t_{i, 2 p + 1}^{G}] \leq s_{i}^{G}, \forall i, p \\ s^{k} [t_{i, 2 p + 2}^{G}] \leq s_{i}^{G} + L_{k}, \forall i, p \\ (a) s^{k} = (L_{k - 1} + d_{0}) 1 + s^{k - 1} + h v^{k} + e^{k - 1} \\ (a') s^{k} = (\sum_{k' \leq k " < k} L_{k "} + d_{0}) 1 + s^{k'} + h v + e^{k'}, k' < k \\ (b) s^{k} = (s_{0}^{k - 1} + L_{k} + d_{0}) 1 + l ν_{0}^{k - 1} + K_{*, 1} a_{0}^{k - 1} + h v + e^{k - 1} \\ (b') s^{k} = (s_{0}^{k - 1} + L_{k} + d_{0}) 1 + l ν_{0}^{k'} + f (a_{0}^{k'}) + h v + e^{k'}, k' < k \\ e^{k'} \geq 0, k' < k \end{cases} \end{array}$
where $D = [\begin{matrix} 1 & - 1 & 0 \\ 0 & 1 & ⋱ \\ ⋮ & ⋱ & - 1 \\ 0 & \dots & 0 & 1 \end{matrix}], L = [\begin{matrix} 0 & 0 \\ 1 & 0 \\ ⋮ & ⋱ \\ 1 & \dots & 1 & 0 \end{matrix}] K = [\begin{matrix} 1 / 2 & 0 \\ 1 + 1 / 2 & 1 / 2 \\ ⋮ & ⋱ \\ k - 1 + 1 / 2 & \dots & 1 + 1 / 2 & 1 / 2 \end{matrix}], I = [\begin{matrix} 1 \\ 2 \\ ⋮ \\ k \end{matrix}]$
, and ${‖ a ‖}_{+}^{2} = {‖ \max (a, 0) ‖}^{2}$
The parameter h is a time headway. The multipliers represent the preferences of each vehicle:

∘ λ ₁ : total acceleration
∘ λ ₂ : absolute speed (safety and emission criteria)
∘ λ ₃ : total travelled distance (time efficiency)
∘ λ ₄ : platooning (throughput maximization)
∘ λ ₅ : maximum acceleration variation (travel comfort)
∘ λ ₆ : platooning with precedent vehicle

The two options (a, a') and (b, b') represent the cases where (a) the full trajectory of the preceding vehicle is known; (b) when only the current position speed and acceleration of the preceding vehicle are known. The function $f (a_{0}^{k})$
is any appropriate function of the input acceleration that represents the evolution of the trajectory of the preceding vehicle and (a') and (b') are generalizations of condition (a) and (b), that could consider all the preceding vehicles for the computation of the current vehicle trajectory. Condition (a) can be substituted by a minimum of the distance to any preceding vehicle trajectory. ^p represents the signal phase of the intersection ⁱ , when multiple phase signal are available.
In Fig. 4 a graphical representation of the speed profile computed by a vehicle according to Fig. 3 is illustrated: A speed profile 1 comprises a two dimensional diagram having a horizontal axis 10 as a time axis and a vertical axis 11 with arbitrary units representing distance s, speed v and acceleration a. The curve 2 illustrates the evolvement of the positions s of a vehicle over time, the curve 3 illustrates the evolvement of the speed or velocity v over time and curve 4 illustrates the evolvement of the acceleration a of the vehicle over time. Further in the speed profile 1 two green light phases GP1 and GP2 are shown and represented by corresponding horizontal lines. Since the speed curve 3 does not reach a value of 0, the vehicle corresponding to the curve 2, 3 and 4 uses the green phases GP1 and GP2 to pass the corresponding traffic light and does not have to stop at the traffic light.
Fig. 5 shows a vehicle V approaching three traffic lights TL1, TL2 and TL3 on its way from left to right illustrating a situation in which a plurality of traffic lights are en route of the vehicle V.
Fig. 6 is illustrating a method according to a second embodiment of the present invention.
In Fig. 6 the three vehicles V1, V2, V3 are shown and the traffic light TL corresponding to the Fig. 1. In Fig. 6 an initiator vehicle, here for example but in general not necessarily the closest vehicle V3 to an intersection with a traffic light TL computes its speed profile 1a based on the information received from the traffic light TL via a transmission TTL. The transmitted information includes the green phase time windows at certain times and lengths denoted in Fig. 6 with reference sign 20. The third vehicle V3 then computes its speed profile 1a based on the green phase information 20 of the traffic light TL and broadcast (reference sign T3) its computed speed profile 1a to the second vehicle V2. For example the second vehicle V2 receives a green phase diagram 20 together with the speed profile 1a including the position curve 2a of the third vehicle V3. The second vehicle V2 then computes its own profile 1b considering green phase information 20 and the speed profile 1a of the third vehicle V3. The second vehicle V2 then broadcasts (reference sign T2) its speed profile 1b together with the green phase diagram 20, the speed profile 1a of the third vehicle V3 and its own speed profile 1b to the first vehicle V1. The first vehicle V1 then computes its own speed profile 1c and uses again the information of the green phase diagram 20 and the previous speed profiles 1a, 1b to compute its own speed profile 1c. The speed profile may then be again broadcasted (reference sign T1) to the other vehicles for further adaption of their speed or operational profiles. This process is continually applied in each vehicle V1, V2, V3 and takes into account the previously transmitted calculated speed profiles 1a, 1b, 1c to compute its own speed profile 1a, 1b, 1c and also considers the green light time windows according to the green light diagram 20. In Fig. 6 the information propagation might be in the opposite direction compared to the driving direction of the three vehicles V1, V2 and V3. Previously received speed profiles 1a, 1b, 1c and traffic light information (green light diagram 20) may be piggybacked by the vehicles V1, V2, V3.
Each vehicle V1, V2, V3 computes its own speed profile 1a, 1b, 1c including the acceleration and based on its current moving state including current speed and position, the distance to other vehicles, especially the vehicle closest to the traffic light TL, the predicted trajectory of surrounding vehicles and the signal time windows communicated by the traffic infrastructure, i.e. in Fig. 6 by the traffic light TL. When calculating its own speed profile the other vehicle speed profiles are considered as constraints like infrastructure information or a distance to the vehicle being adjacent to the traffic light.
Fig. 7 is illustrating signal phase adjustments based on speed profiles.
In Fig. 7 signal phase adjustments operated by a traffic light controller for a traffic light based on the received speed profiles are performed. The traffic light TL or its corresponding controller receives for example all speed profiles 1a, 1b, 1c aggregated or in individual messages and adapts the time windows of the green phases GP1, GP2 accordingly.
The traffic light or more general the signaled infrastructure adapts the green phase time windows GP1, GP2 based on the speed profile 1a of the last vehicle V1 of the vehicle platoon V1, V2, V3 that is approaching the traffic light TL. The traffic infrastructure may dynamically choose which vehicle V1, V2, V3 will be the last vehicle allowed to pass in the green phase time window based on the received speed profiles 1a, 1b, 1c and the number of vehicles waiting for a green phase time window in other intersecting roads (ex-route information). The decision may also be based on external information for example presence of vehicles in other conflicting directions (ex-route-information) or by energy or emission considerations (preference information) in particular in a vehicle platoon. In Fig. 7 the first green phase GP1 is for example reduced, since the trajectories 2a, 2b and 2c of the vehicles V1, V2, V3 pass the traffic light TL in green phase time window GP1 in the second half of the first green phase time window GP1. Therefore the beginning of the green phase time window GP1 may be shifted to a later time allowing vehicles in other intersecting roads to be provided with longer green phase time windows. The first green phase GP1 is amended in such a way, that the trajectories 2a, 2b and 2c may still each fall into the first green phase time window GP1. Accordingly the second green phase time window GP2 at a second intersection may also be amended. In Fig. 7 the end of the second green phase time window GP2 may be reduced, since the vehicle platoon V1, V2, V3 passes the corresponding traffic light in the first half of the green phase time window GP2, therefore the green phase time window GP2 may be reduced in time length.
Fig. 8 is illustrating a method according to a third embodiment of the present invention.
In Fig. 8 the traffic light TL restarts an iterative process to obtain new speed profiles of the vehicle platoon V1, V2, V3 based on a new green phase timing, for example as described in Fig. 7. This negotiation process aims at reducing, enlarging and/or shifting the green phase time windows for multiple purposes, as for example to improve overall energy efficiency of the vehicle platoon or a traffic throughput maximization at the intersection with the traffic light TL. The traffic light TL sends its green phase information 20 to the vehicles V1, V2, V3. The vehicles V1, V2, V3 compute their speed profiles 1a, 1b, 1c and send them back to the traffic light TL. The traffic light TL then recomputes its green phase time windows and sends the new green phase information back to the vehicles V1, V2, V3. The vehicles V1, V2, V3 then recomputed again their speed profiles 1a, 1b, 1c based on the new green phase information 20 of the traffic light TL and so on.
In each iteration step, the new proposed green phase timing of the traffic light TL may be validated by each vehicle V1, V2, V3 of the vehicle platoon V1, V2, V3. At the end of each successful negotiation iteration, the traffic light TL confirms the accepted new green phase. This negotiation process may continue until one, some or all of the vehicles of the platoon do accept the new proposed green phase timing. It is further possible to shorten the red phase of the traffic light TL for the part not used by the vehicle trajectories 2a, 2b, 2c plus some safety margin. It is also possible that the traffic light TL may also force some vehicles to stop or to enlarge the green phase time window to let stopping vehicles to pass.
Fig. 9 is illustrating a fourth embodiment of the present invention.
In Fig. 9 an iterative process between different vehicles V, V1, V2, V3 is shown. A vehicle V of the vehicle platoon V, V1, V2, V3 restarts in contrast to Fig. 8 the iteration in order to further optimize the overall platoon's speed profile based on the information received from the other vehicles V1, V2, V3 of the vehicle platoon. During the iterations, each vehicle V, V1, V2, V3 modifies its own speed profile based on the aggregated speed profiles of all vehicles V, V1, V2, V3 in the vehicle platoon. For each vehicle V, V1, V2, V3, the last received profile is valid until a new one is transmitted. This means that it is not necessary to complete the full iteration. A traffic light TL may use the last received profile to adjust the traffic light phase including the green phase time windows.
In Fig. 9 vehicle V computes its own speed profile and transmits its speed profile to the other vehicles V1, V2, V3 of the vehicle platoon. The other vehicles V1, V2, V3 compute their own speed profiles based on the transmitted speed profile of vehicle V and transmit their speed profiles back to the vehicle V. The vehicle V evaluates, if improvement margins are provided. If improvement margins are provided the vehicle V recomputes its own speed profile based on the improvement margins and sends its speed profile back to the other vehicles V1, V2, V3 of the vehicle platoon. The other vehicles V1, V2, V3 then recompute their speed profiles based on the improved speed profile of vehicle V and so on. It is further possible to foresee that the vehicle changes its preferences, and adds then a preceding vehicle trajectory as constraints in the opposite way, wherein the current vehicle shall precede a following vehicle and checks after that, whether there is any advantage for the following vehicle, for example more green phase time left for the following vehicle to pass the intersection.
Fig. 10 is illustrating an effect of uncertainties in a vehicle's trajectory.
In Fig. 10 an evolvement of a vehicle trajectory 21, 22, 23, 24 and 25 is shown over time with small variations in a current moving state of the vehicle. Further variances VAR1, VAR 2 and VAR3 are shown which define the evolvement of the variance in the different trajectories 21, 22, 23, 24, 25 in the future. Exchanging information among vehicles including information on specific constraints like minimum and/or maximum acceleration or preferences in the trajectory computation enables to extent the calculation of the speed profiles beyond the nominal trajectory by considering a family of trajectories with different specific weight or probability. The exchange information is used to calculate trajectories which are more robust to local perturbations, for example speed fluctuation due to more or less wind, etc. Fig. 10 illustrates such effects of uncertainty in the future trajectories of vehicles.
Fig. 11 is illustrating effects of different weight choice.
In Fig. 11 the effect of different weight choice according to the formula (1) of Fig. 3 is shown. Different choice of weighting parameters in the calculation results in different trajectories 21, 22 representing total travelled distance s' and s. Of course this results also in different total speed and different total acceleration. $\begin{matrix} {‖ a ‖}_{+}^{2} < {‖ a' ‖}_{+}^{2} \\ {‖ v ‖}^{2} < {‖ v' ‖}^{2} \\ s' [T] > s [T] \end{matrix}$
As shown in Fig. 11 the different green phase time windows GP1 and GP2 have to be adapted accordingly otherwise the vehicle with the trajectory 21, 22 would have to stop.
Fig. 12 is illustrating a signal phase split.
In Fig. 12 a signal phase split to accommodate multiple vehicles is shown. The overall green phase time window GP is split into single slots GP1, GP2, GP3 and GP4 which are allocated to each vehicle. t^G _i,1 and t^G _i,2 represent start and end time of the green phase time window GP at an i-th traffic light at the i-th intersection.
Fig. 13 is illustrating signal phase start and end time with and without safety time margins. In Fig. 13 t^G _i,1 and t^G _i,2 represent start and end time of the green phase time window GP at an i-th traffic light at the i-th intersection. This is shown on the upper line in Fig. 13. The line below includes when changing from red phase to the green phase GP' safety margins SM1 and SM2 for safety requirement which may be implemented as yellow signal phase in traffic lights. When applying equation (1) for calculating the speed profiles based on the green phase time windows, a feasibility step should also be performed where the green phase time windows which are more appropriate are selected and when infeasibility is detected, unreachable green phase time windows are removed.
Fig. 14 is illustrating a partial common driving path of two vehicles.
Fig. 14 shows a complex intersection model: En route of a first vehicle V1 are three intersections IS1, IS2 and IS3 and corresponding traffic lights TL1, TL2 and TL3 regulating the vehicle flow at the three intersections IS1, IS2 and IS3. En route of a second vehicle V2 are the second and third intersection IS2 and IS3. Since the second vehicle is approaching the second intersection from another direction compared to the first vehicle V1 the second vehicle passes traffic light TL2' at the second intersection IS2 and at the third intersection IS3 the same traffic light TL3 as the first vehicle V1. The two vehicles V1, V2 therefore only partially share the driving path represented by reference sign A in Fig. 14 between the second and third intersection . Of course, constraints according to the method of the present invention may be considered only of the common path section A. The traffic lights TL1, TL2 and TL3 are en route of the first vehicle V1, the traffic lights TL2' and TL3 en route of the second vehicle. At each intersection IS1, IS2, IS3 a traffic light signal controller TFRD1, TFRD2 and TFRD3 controls the corresponding traffic lights TL1, TL1', TL2, TL2' and TL3 for the corresponding intersection, e.g. at the first intersection IS1, traffic light signal controller TFRD1 controls the traffic lights TL1, TL1'.
Fig. 15 is illustrating a system according to a fifth embodiment of the present invention.
In Fig. 15 an example implementation of a system according to the invention is shown. A first module M1 computes the optimal speed in scenarios, whether there are no vehicles in front, i.e. current and past distances and speed differences to preceding vehicles are not used as input parameter. The second module M2 uses the input parameters of module M1 and further the current and past distances and speed differences to a preceding vehicle. The second module M2 may be connected to radar sensors to sense speed and acceleration, distance and position of a preceding vehicle keeping platooning distance and safety distance. Of course, this information may also be obtained by direct communication with the preceding vehicle. Module M2 provides as output parameters the future position, speed and acceleration of the current vehicle. The third module M3 then computes future position, speed and acceleration of the current vehicle and if applicable based on the current and future position of the preceding vehicle. The first module M1 could be implemented by using equation (1) and using MPC while the other two modules M2, M3 may be implemented based on fuzzy logic control (FLC).
It is also possible to calculate the green phase time windows proposed by a traffic controller respectively traffic infrastructure by taking into account both the previous iteration of speed profiles exchanges among vehicles, for example according to Fig. 9 and a further iterative exchange of green phase time windows among adjacent further traffic lights to further enhance for example traffic flow.
Fig. 16 is illustrating a flow chart according to a sixth embodiment of the present invention. After starting (step S1) signal phase information and preceding and/or following vehicle speed profiles are received in a second step S2. In a third step S3 the vehicle computes its own speed profile which is used to set for example the current speed. The output of the computation, i.e. the computed speed profile is then evaluated in a forth step S4 for improvement margins. If there are improvement margins the speed profile is recalculated and rechecked for improvement margins. If there are no improvement margins the computed speed profile is then communicated or transmitted to the other vehicles and the traffic infrastructure, for example a traffic light: In a fifth step S5 a vehicle controller is initiated which decides in a sixth step S6 if feedback for the other vehicles and/or traffic infrastructure is needed. If needed in a seventh step S7 feedback is sent to other vehicles and/or to an intersection signal, for example a traffic light. In an eighth step S8 the computed improved speed profile is sent to the other vehicles and/or the traffic light. In a ninth step S9 it is decided if the steps S2 to S9 should be performed again, for example when receiving additional external information provided to the vehicle. If there are no external information, the process is stopped in a tenth step S10. The steps S2 to S9 are performed iteratively over time due to changes in the vehicle moving states and moving profiles and possible external information provided to the vehicle.
Fig. 17 is illustrating a flow chart according to a seventh embodiment of the present invention. In Fig. 17 a traffic light controller flow chart is shown. After starting (step S1) signal phase information and preceding and/or following vehicle speed profiles are received in a second step S2. An improved signal phase in a step S3 is then computed based on the received information. In a fourth step S4 it is decided if the computed signal phases of the traffic light provides still an improvement margin. Then the step S3 and again step S4 is repeated. If there is no improvement margin anymore this information is transmitted in a fifth step S5 to a corresponding controller unit in the traffic light controller. The traffic light controller then sends the improved signal phase in a step S6' to the vehicles. In a seventh step S7' is decided to repeat the steps S2-S7'. If not in an eighth step S8' the steps S2' to S7' are stopped otherwise steps S2-S7' are performed.
The traffic light controller updates according to the steps S2 to S7' its signal phase based on receiving speed profiles from approaching vehicles and/or surrounding further traffic lights or intersection controllers. The step S2 to S7' may be also performed iteratively. The iteration according to Fig. 17 as well as Fig. 16 is not to be considered sequential. In particular, the last vehicle may compute its own speed profile without waiting for other vehicles.
In summary the present invention enables smooth trajectories for vehicles passing signaled intersections and also negotiations of green phases within a vehicle platoon and a signal intersection itself. The present invention further provides an integration of speed profiles from multiple vehicles to compute a vehicle speed profile and traffic light phases. The present invention further provides nested iterative negotiation processes for traffic optimization including traffic light phase optimization and speed profile optimization within a vehicle's platoon while ensuring respectively maintaining vehicle platooning requirements regarding safety, vehicle characteristics, energy efficiency and comfort while optimizing intersection traffic efficiency.
The present invention further provides an integration of traffic infrastructure information for a cruise control system for smooth speed advice, integrates different information origins from a distance determining system for vehicles with information coming form broadcasted messages. The present invention further provides an integration of the information from preceding and following vehicles to provide a platooning solution. The present invention even further provides an interaction with signaled intersections and interaction with other vehicles and enables an optimization of signal phases of traffic lights based on a negotiation with vehicles and other intersections.
The present invention provides also an iterative mechanism that does not cause stability issues due to incomplete iteration cycles. Iterations among the vehicles and the intersection controllers and among the intersection's controllers proceed only when the modification introduces a positive effect, measured by the intersection controller using an objective function, based for example on better throughput, less energy, less overall travel time. The iteration on a specific green phase time window can also consider safety margins, as for example when a green phase has started, the start time cannot be modified anymore. Conventional iterative methods suffer ripple phenomenon, when the phase start and end values oscillate close to the minimum, then the iteration step may be reduced when approaching a minimum point of the objective function. Iterative process on single green phase can be stopped at any time by the intersection controller; in this case, the last identified solution is applied by the traffic controller.
Many modifications and other embodiments of the invention set forth herein will come to mind the one skilled in the art to which the invention pertains having the benefit of the teachings presented in the foregoing description and the associated drawings.

Claims

A method for adapting vehicular traffic flow with a plurality of vehicles (V1, V2, V3) and at least one traffic flow regulation device (TFRD1, TFRD2, TFRD3) outside the vehicles (V1, V2, V3),
characterized by the steps of
a) Determining current moving states (1a, 1b, 1c) of the vehicles (V1, V2, V3), wherein a current moving state (1a, 1b, 1c) includes driving condition information of a vehicle (V1, V2, V3),

b) Determining a current device state of at least one of traffic flow regulation devices (TFRD1, TFRD2, TFRD3) wherein a device state includes moving profile impact information for the vehicles (V1, V2, V3), wherein moving profile impact information represents information leading to a possible amendment of the moving profile and/or the moving state,

c) Calculating moving profiles (1a, 1b, 1c) of the vehicles (V1, V2, V3), wherein a moving profile represents current and/or future moving states and/or calculating an operational profile (20) of at least one of the traffic flow regulation devices (TFRD1, TFRD2, TFRD3), wherein an operational profile (20) represents current and/or future device states based on the corresponding current moving states (1a, 1b, 1c) and device states

d) Exchanging (T1, T2, T3, TTL) profile information of the calculated moving profiles (1a, 1b, 1c) and possibly additionally the calculated operational profile, between a traffic flow regulation device (TFRD1, TFRD2, TFRD3) and vehicles (V1, V2, V3)

e) Adapting the moving profiles (1a, 1b, 1c) and/or the operational profile (20) according to the exchanged profile information of step d)

wherein at least steps d) and e) are performed iteratively until a predetermined convergence criterion is fulfilled.
The method according to claim 1, characterized in that an operational profile (20) and/or moving profile impact information include signal phase information (GP1, GP2), preferably of at least one traffic light signal controller for traffic lights (TL1, TL2, TL3), for a vehicle (V1, V2, V3).
The method according to one of the claims 1-2, characterized in that the driving condition information for a vehicle (V1, V2, V3) include a speed profile (1a, 1b, 1c) and/or relative distances (D12, D23) to preferably directly adjacent vehicles (V1, V2; V2, V3).
The method according to one of the claims 1-3, characterized in that the steps a)-e) are initiated by a vehicle (V3) of the plurality of vehicles (V1, V2, V3), preferably the vehicle (V3) which is the closest to the traffic flow regulation device (TFRD1, TFRD2, TFRD3).
The method according to one of the claims 1-4, characterized in that profile information of the current moving profile (1a, 1b, 1c) is received by the traffic flow regulation device (TFRD1, TFRD2, TFRD3) and its operational profile (20) is adapted according to the received information.
The method according to one of the claims 1-5, characterized in that the operational profile (20) of the traffic flow regulation device (TFRD1, TFRD2, TFRD3) is adapted according to ex-route information, preferably including a number of vehicles waiting for a green phase time window at other intersecting roads and/or a presence of vehicles in other conflicting directions.
The method according to one of the claims 1-6, characterized in that the operational profile (20) of the traffic flow regulation device (TFRD1, TFRD2, TFRD3) is adapted according to external physical information, preferably including energy and/or emission considerations in certain urban areas and/or a number of vehicles to pass a traffic light.
The method according to one of the claims 1-7, characterized in that an iteration of at least the steps d)-e) is again performed for the moving profile (1a, 1b, 1c) when an operational profile (20) was adapted.
The method according to one of the claims 1-8, characterized in that an iteration of at least the steps d)-e) is again performed for the moving profiles (1a, 1b, 1c) and between the vehicles (V1, V2, V3) only, preferably wherein
a result of the iteration is transmitted to the traffic flow regulation device (TFRD1, TFRD2, TFRD3) and the operational profile (20) of the traffic flow regulation device (TFRD1, TFRD2, TFRD3) is adapted according to the transmitted result.
The method according to one of the claims 1-9, characterized in that in step d) constraint information and/or preference information for the states (1a, 1b, 1c, 20) are exchanged, preferably wherein
the constraint information and/or the preference information for the states (1a, 1b, 1c, 20) are weighed.
The method according to one of the claims 3-10, characterized in that the moving profile (1a, 1b, 1c), preferably the speed profile of a vehicle (V1, V2, V3) is calculated based on total acceleration, absolute speed, total travelled distance, platooning of vehicles (V1, V2, V3), maximum acceleration variation and/or platooning with preceding vehicles (V1, V2, V3), preferably wherein
the moving profile (1a, 1b, 1c), preferably the speed profile, is either calculated based on at least partially known moving profile (1a, 1b, 1c), preferably at least position, speed and acceleration, of a preceding vehicle (V1, V2; V2, V3).
The method according to one of the claims 1-11, characterized in that a time slot (GP1, GP2, GP3, GP4) is assigned to each vehicle (V1, V2, V3) in the traffic flow regulation device (TFRD1, TFRD2, TFRD3), preferably wherein
the time slot (GP1, GP2, GP3, GP4) includes a safety time margin (SM1, SM2).
The method according to one of the claims 1-12, characterized in that the operational profile (20) of the traffic flow regulation device (TFRD1, TFRD2, TFRD3) is calculated based on a previous complete iteration having fulfilled the convergence criterion and an iteration of exchanging and adapting operational profiles (20) of adjacent traffic flow regulations devices.
A system for adapting vehicular traffic flow with a plurality of vehicles (V1, V2, V3) and at least one traffic flow regulation device (TFRD1, TFRD2, TFRD3) outside the vehicles (V1, V2, V3), for performing a method according to one of the claims 1-13, characterized by
at least two moving state determining means for determining a current moving state (1a, 1b, 1c) of a vehicle (V1, V2, V3), wherein a current moving state (1a, 1b, 1c) includes driving condition information of a vehicle (V1, V2, V3),

device state determining means for determining a current device state of at least one of the traffic flow regulation devices (TFRD1, TFRD2, TFRD3), wherein a device state includes moving profile impact information for the vehicles (V1, V2, V3), wherein moving profile impact information represents information leading to a possible amendment of the moving profile,

moving profile calculation means for calculating a moving profile (1a, 1b, 1c) for a vehicle (V1, V2, V3) , wherein a moving profile represents current and/or future moving states and/or operational profile calculation means for calculating an operational profile (20) of at least one of the traffic flow regulation devices (TFRD1, TFRD2, TFRD3), wherein an operational profile (20) represents current and/or future device states based on the corresponding current moving states (1a, 1b, 1c) and device states, transmission and receiving means for exchanging profile information of the calculated moving profiles (1a, 1b, 1c) and/or the calculated operational profiles (20) between a traffic flow regulation device (TFRD1, TFRD2, TFRD3) and vehicles (V1, V2, V3) and adaption means for adapting the moving profiles (1a, 1b, 1c) and/or the operational profile (20) according to the exchanged profile information, wherein the transmission and receiving means and the adaption means are operable to perform exchanging and adapting the moving profiles (1a, 1b, 1c) and possibly additionally the operational profiles (20) iteratively until a predetermined convergence criterion is fulfilled.
The system according to claim 14, characterized in that the moving profile calculation means are provided in three modules, wherein
the first module (M1) is operable to perform the steps a) to e) of claim 1 for a vehicle (V3) being adjacent to the traffic flow regulation device (TFRD1, TFRD2, TFRD3), the second module (M2) is operable to perform the steps a) to e) for a vehicle (V1, V2) having a preceding vehicle (V2, V3) in front, and the third module (M3) is operable to combine results from the first and second module (M1, M2), preferably wherein

the first module (M1) is operable to perform model predictive control (MPC) and the second and third module (M2, M3) are operable to perform fuzzy logic (FLC).