Driving assistance method and in-vehicle driving assistance system
Patent Information
- Application Number
- JP2022567412
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2020-05-05
- Filing Date
- 2021-04-28
- Publication Date
- 2024-11-14
- Estimated Expiration
- 2041-04-28
Smart Images

Figure 0007583830000027 
Figure 0007583830000028 
Figure 0007583830000029
Abstract
Description
[Technical field]
[0001] The present invention relates generally to the automotive field, and more specifically to driving assistance for automobiles. [Background technology]
[0002] To increase road safety, certain automobiles, called semi-autonomous automobiles, are equipped with partially automated systems or advanced driver assistance systems (known by the acronym ADAS), in particular systems that perform the lateral and / or longitudinal control of the vehicle instead of the driver, or at least warn the driver of potentially dangerous situations to enable him to react in time. It is also possible to make automobiles fully autonomous, i.e. without a driver.
[0003] In order to enable an autonomous or semi-autonomous vehicle (hereinafter referred to as the "target vehicle") to detect dangerous situations and react accordingly to avoid or reduce the risk of an accident, the vehicle's on-board driving assistance system must not only be able to detect all dynamic objects (hereinafter referred to as "third-party vehicles") present in the vehicle's current environment, such as other motor vehicles (cars, trucks, motorcycles), but also be able to predict the future movements of these third-party vehicles.
[0004] As described, for example, in the article entitled “A survey on motion prediction and risk assessment for intelligent vehicles” (Lefevre et al., Robomech Journal 2014.1:1 http: / / www.robometechjournal.com / content / 1 / 1 / 1), known trajectory prediction methods are based on a motion model selected from among three types of motion models: -Physics-based movement model, - A maneuver-based motion model, and -Interaction-sensitive motion model.
[0005] Physics-based motion models are the simplest, as they consider the future motion of the vehicle to depend only on the laws of physics. These models rely heavily on the accuracy of the on-board sensors of the target vehicle and assume that the third-party vehicle does not change its state (speed or direction). As a result, these models do not allow to reliably predict the trajectory of the third-party vehicle for long-term predictions, e.g., beyond 2 seconds.
[0006] The maneuver-based motion model is a bit more sophisticated than the physics-based motion model because it also takes into account the maneuvers that the driver of the third party vehicle intends to perform. Nevertheless, the target vehicle and the third party vehicle, whose maneuvers are estimated, are considered to be moving independently of each other, which may lead to a misinterpretation of certain road conditions and affect the risk assessment of the target vehicle.
[0007] Interaction-sensitive motion models are currently the most advanced, since they take into account the fact that the motion of a vehicle may be influenced by the motion of other vehicles present in the road scene. Most of these models use dynamic Bayesian networks, which allow taking into account pairwise dependencies between several moving vehicles. Although these models allow for reliable and longer-term predictions, they require significant computation time resources to be able to estimate all potential trajectories of vehicles for all possible pairs of vehicles, and are therefore incompatible with the need to assess the risk to the target vehicle in real time. [Prior art documents] [Non-patent literature]
[0008] [Non-Patent Document 1] Lefevre et al., Robomech Journal 2014.1:1 http: / / www.robometechjournal.com / content / 1 / 1 / 1 Summary of the Invention
[0009] One object of the present invention is to overcome the limitations of the prior art, in particular by proposing a simplified method that makes it possible to predict reliably and cheaply in terms of computation time at least one future position of each vehicle present in a road scene in which a target vehicle is moving.
[0010] Another object of the present invention is to use this simplified method to rapidly predict the trajectories of multiple vehicles moving within an environment of third party vehicles over a longer prediction time than methods that use physics-based motion models.
[0011] Thus, one subject of the present invention is A driving assistance method for a target automobile vehicle, the method being implemented in an in-vehicle driving assistance system on the target automobile vehicle, comprising: - a detection step in which the in-vehicle driver assistance system on the target vehicle detects a number of third party vehicles that are initially present in the environment of the target vehicle in a multi-lane driving zone; - for each vehicle O in the set, for a first predicted time point subsequent to the initial time point in a first coordinate system associated with the target vehicle by the in-vehicle driving assistance system; k a first prediction cycle of prediction of the first relative position and the first relative velocity; wherein the first prediction cycle comprises: Each vehicle in a set of K vehicles including the third party vehicle and the target automobile vehicle detected at the initial time point k storing initial data of each vehicle O in the set in the first coordinate system at the initial time in a database of the in-vehicle driving assistance system, k a current lane, a current relative position, and a current relative speed of the vehicle; 〇 The K vehicles in the set O k a sorting step in which a priority is assigned to each vehicle O in the set stored in the database, k and the lane, and starting from the vehicle detected at the furthest forward position of the target motor vehicle, k a sorting step, corresponding to the order in which Each vehicle in the set in the order of priority k A selection step of selecting For each selected vehicle O* in the set k In contrast, said selected vehicle O* k a first sub-step of identifying another vehicle in the set that may be a primary target vehicle for the identified primary target vehicle, the selected vehicle O* k and the selected vehicle O* k based at least on information about the current environment of the selected vehicle O* k a second sub-step of estimating a maneuver that is in progress or is about to be performed by said selected vehicle O* k Based on the maneuver estimated for the selected vehicle O* between the initial time point and the first predicted time point, k a third sub-step of estimating the movement of Including, The selected vehicle O* for the first prediction time point k the first relative position and the first relative velocity of the vehicle are predicted based on the estimated movement; The first sub-step, the second sub-step, and the third sub-step are k based on position and velocity data transformed into a two-dimensional coordinate system associated with A driving assistance method.
[0012] In one possible embodiment, the estimated maneuver is selected from a predefined set of possible maneuvers.
[0013] In one possible embodiment, the selected vehicle O* k said predefined set of possible maneuvers for said selected vehicle O* k maintain current lanes; said selected vehicle O* k change lanes to the left, said selected vehicle O* k change lanes to the right, said selected vehicle O* k is the selected vehicle O* k Stop in your current lane, Includes.
[0014] In one possible embodiment, the first substep comprises: k The method includes searching for a vehicle in the set that is in front of and closest to the vehicle in the same lane.
[0015] In one possible embodiment, the selected vehicle O* k The information relating only to the selected vehicle O* k and / or the lateral movement of the selected vehicle O* k and / or the on or off state of one of the indicators of the selected vehicle O*. k Contains a history of stored locations.
[0016] In one possible embodiment, the selected vehicle O* k The information about the current environment of the selected vehicle O* k the type of markings of the lane in which the selected vehicle O* is located, and / or k the occupancy status of lanes adjacent to said current lane, and / or the current speed limit assigned to said lane.
[0017] In one possible embodiment, the method comprises: calculating, by the in-vehicle driver assistance system, a time point for each vehicle O in the set for a second prediction time point subsequent to the first prediction time point in the first coordinate system. k The second relative position {X;Y;θ} k,2 and the second relative velocity
number
number
[0018] In one possible embodiment, the method includes a number N of successive prediction cycles, the number N being greater than 2, each n-th prediction cycle being a prediction cycle for each vehicle O in the set in the first coordinate system and for an n-th prediction time instant following a preceding prediction time instant. k nth relative position {X;Y;θ} k,n and the nth relative velocity
number
[0019] In one possible embodiment, successive prediction time points are separated by a constant time step.
[0020] In one possible embodiment, N is equal to 33 and the fixed time step is equal to 200 ms.
[0021] Another subject of the invention is an on-board driving assistance system on the subject motor vehicle, which implements the method according to the invention.
[0022] The invention will be better understood in view of the following description, presented with reference to the accompanying drawings. [Brief description of the drawings]
[0023] [Figure 1] 1 shows, in plan view, a schematic representation of an example road scene useful for illustrating the principles of the invention; [Diagram 2]For the example road scene of FIG. 1, the trajectory predicted by the method according to the invention is shown diagrammatically. [Diagram 3] 2 shows the steps that can be carried out according to a first embodiment of the detection method according to the invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0024] To present a concrete concept, the invention will now be described in the context of the non-limiting example of a road scene, shown diagrammatically in plan view in FIG.
[0025] In this FIG. 1, a target vehicle O having an advanced driver assistance system (not shown) according to the present invention is shown. I The target vehicle O is moving in a driving zone having three lanes L1, L2, and L3. The other five vehicles O1 to O3, O5, and O6 are also moving in the driving zone having three lanes L1, L2, and L3. I The subject vehicle is moving in an environment of a vehicle that is not in a limiting manner. For simplicity, it is assumed in a non-limiting manner that the driving zone corresponds to a part of a highway with two lanes L2, L3, lane L1 corresponds to a feeder road, and all vehicles are moving in the same direction (from left to right in FIG. 1 ) according to French traffic rules (overtaking from the left and speed limited to 130 km / h). In a non-limiting example, the third party vehicles O1, O5, and O6 are all cars, the third party vehicle O2 is a motorcycle, and the third party vehicle O3 is a truck. Furthermore, in this example, the truck O3 is stationary and the brake lights of the third party vehicle O1 are on because its driver is braking. The nature of the third party vehicles present in the environment of the subject vehicle does not affect the principles of the present invention. In other words, the third party vehicles may be either conventional vehicles, semi-autonomous vehicles, or autonomous vehicles.
[0026] In the following, the target vehicle O I teeth, On the other hand, the target vehicle O I On the other hand, information about the geometric shape of the road scene (in particular, marking lines, traffic signs, etc.) is provided to the target vehicle O Ivarious sensors (not shown) of different nature (e.g. image sensors, radar, lidar) that allow the -Process the detected information and make appropriate decisions (target vehicle O I Warning to the driver of the vehicle and / or I It is assumed that the vehicle is equipped with an ADAS (not shown) that enables the vehicle to perform lateral and longitudinal control of the vehicle.
[0027] The complete driving assistance method according to the present invention includes detecting the presence of various third party vehicles at an initial time t0 and detecting the presence of the target vehicle O I and predicting the future trajectories (or predicted trajectories) for all third party vehicles whose presence was detected at the initial time instant t0 over a given overall prediction period.
[0028] In what follows, the following notations are used: -O k is the target vehicle O I 1, the set represents a total of six vehicles, i.e., the detected third party vehicles O1-O3 and O5, O6, as well as the following criteria O I Or the target vehicle O that can replace O4 I Including, -P k (0) is the vehicle O in the set of K vehicles at the initial detection time t0. k represents the point associated with this point P k (0) is the conventional method, Target vehicle O I (see FIG. 1 ) k current relative position {X;Y;θ} k,0 (Here, θ is the coordinate system of the first coordinate system O I Vehicle O k radians), and Target vehicle O I Vehicle O in this same first two-dimensional coordinate system k The current relative velocity of
number
number
number
[0029] Therefore, each vehicle O obtained from the set including the detected third party vehicle and the target vehicle k The predicted trajectory for is the initial point P measured at the initial time t0. k Starting from (0), N points P are successively estimated at time steps Δt where n goes from 1 to N. k It is formed by a sequence of (n).
[0030] In one embodiment, the time step Δt is constant between successive points of the predicted trajectory. By way of example, a time step Δt equal to 200 ms is chosen and the number N is set equal to 33, so that each vehicle O is predicted for a total prediction time of 7 seconds, starting from the initial time t0. k FIG. 2 shows a trajectory prediction for various vehicles O according to the principles of the present invention, which are described in detail below. k 1 shows an example of various predicted trajectory portions for a vehicle O5. For example, the predicted trajectory for a vehicle O5 includes an initial point starting at point P5(0), a point P5(5) at prediction time t5, and a point P5(6) at prediction time t 10 The predicted trajectory for vehicle O6 includes a series of predicted points, such as point P5(10) at prediction time t, and a series of predicted points, such as point P6(0) at prediction time t, and point P6(7) at prediction time t. 19 and a series of predicted points such as point P5(19) in
[0031] Referring to FIG. 3, a method 100 according to the present invention includes: I (O4) on-board system, each vehicle O k Regarding point P k Based on (0), the first point P k A method 100 is described that includes at least one first prediction cycle that allows predicting (1).
[0032] The method 100 includes an initial detection step 110 during which a target motor vehicle O I The above in-vehicle system detects a target vehicle O at an initial time t0 in a multi-lane driving zone. I Therefore, in the example of FIG. 1 and FIG. 2, the target vehicle O I Assume that the above in-vehicle system detects the presence of third party vehicles O1-O3, O5, and O6 at an initial time t0.
[0033] According to the notation above, the purpose of this first prediction cycle is to predict the target vehicle O I The above in-vehicle system is at the first point P k(1), i.e., for each vehicle O in the coordinate system associated with the target vehicle and for a first prediction time t1 following the initial time t0, k The first relative position {X;Y;θ} k,1 and the first relative velocity
number
[0034] For this purpose, the first prediction cycle is performed by detecting the third party vehicle and the target vehicle O detected at the initial time t0. I For each vehicle O in a set of K vehicles including k The process starts with step 120, in which the initial data is stored in the database of the in-vehicle system. -On the other hand, for each initial point P k (0), i.e., for each vehicle O in the set at the initial time t0 k current relative position {X;Y;θ} k,0 and the current relative velocity
number
[0035] Table 1 below gives an example of the content of the database at the detection time t0 of the road scene shown in FIGS. [Table 1]
[0036] According to an important feature of the invention, the first prediction cycle is performed for K vehicles in the set O k Following a sorting step 130 in which a priority is assigned to each vehicle O in the set stored in the database, the priority is k and the target vehicle O I Vehicles in the set O in the driving zone starting from the vehicle detected at the furthest forward position of k In the case of the road scene presented as an example in Figures 1 and 2, the truck O3 appears as the first vehicle to be considered, followed by the motorcycle O2, then the vehicle O1, the target vehicle O4, the vehicle O5 and finally the vehicle O6, as indicated by the numbers placed on the rear of each vehicle.
[0037] Table 2 below shows Table 1 with its columns sorted according to priority corresponding to the example road scene of FIGS. [Table 2]
[0038] The first prediction cycle is performed by the on-board system of the target vehicle to predict the vehicle O k Each of the following (target vehicle O I More specifically, the target vehicle O I The in-vehicle system assigns priority to each vehicle O in step 130. k (step 140). Each vehicle selected in priority order is then k 1 and 2, a separate prediction process is first performed for the selected vehicle O3* corresponding to the truck O3, then for the selected vehicle O2* corresponding to the motorcycle O2, and so on until the last vehicle O6* is selected from table 2 corresponding to the vehicle O6.
[0039] For each selected vehicle k The individual processing performed on * essentially involves the following sub-steps, which are described in more detail below: - Selected vehicle O* k substep 150 of identifying another vehicle in the set that may be a primary target vehicle for - the primary target vehicle identified in substep 150, the selected vehicle O* k Information about only selected vehicles O* k Based on at least information about the current environment of the selected vehicle O* k a substep 160 of estimating a maneuver that is in progress or is about to be executed by - Selected vehicle O* k Based on the maneuver estimated in substep 160 for the selected vehicle O* between the initial time t0 and the first predicted time t1. k Substep 170 of estimating the movement of
[0040] To simplify the calculations, sub-steps 150, 160, and 170 are performed by first calculating the target vehicle O I Data expressed in a first coordinate system associated with the vehicle O, which is the vehicle selected for processing. I (except in relation to the vehicle selected) k Therefore, all prediction processes are performed based on data transformed into a two-dimensional coordinate system associated with the target vehicle O. I The above in-vehicle system was actually selected for vehicle O* k The system is therefore executed as if it were installed in the selected vehicle O* k For each prediction process related to the selected vehicle O*, all the data stored in the sorted initial database is k , and this information is transformed in advance (by rotation and translation) into the coordinate system associated with the selected vehicle O* k must be stored in a temporary database for display.
[0041] As an example, Table 3 below shows the vehicle O* selected for processing by substeps 150-170. k corresponds to truck O3 (first processing according to the priority order), and the following Table 4 shows the vehicle O* selected for processing according to sub-steps 150 to 170. k 1 presents a temporary database obtained when serves vehicle O3 (second process according to priority). [Table 3] [Table 4]
[0042] In the above Tables 3 and 4, -{X;Y;θ}* k,0 is the selected vehicle O* k Each vehicle O after being transformed into a two-dimensional coordinate system associated with k represents the current relative position of -
number
[0043] Based on the transformed information stored in the temporary database representing the selected vehicles, the in-vehicle system selects a vehicle O* in the set of K vehicles. k (substep 150 above). Such potential primary targets can be identified in a conventional manner by the selected vehicle O*. k According to the converted information stored in the temporary database, the selected vehicle O* kThe vehicle in question is identified by the in-vehicle system of the vehicle in question by searching for the vehicle in the set that is in front of and closest to the selected vehicle O* in the same lane. This search is performed in a conventional manner by searching for the vehicle in question O* k This is done by searching for the vehicle with the shortest estimated time to collision (i.e., TTC) with the
[0044] In the example road scene shown in Figures 1 and 2, during the sub-step 150 carried out for the selected vehicles corresponding to the truck O3, the motorcycle O2 or the vehicle O6, no primary target will be detected since no vehicle is located directly ahead of these vehicles and in the same lane. On the other hand, the primary target that will be detected at the end of the sub-step 150 is truck O3, if the vehicle selected for processing corresponds to vehicle O1; vehicle O1, if the vehicle selected for processing corresponds to the target vehicle O4, and - The vehicle selected for processing is the motorcycle O2, where the vehicle corresponds to the target vehicle O5.
[0045] The in-vehicle system then determines the selected vehicle O* using, inter alia: k This allows for an estimation of the maneuver that is in progress or about to be executed (substep 160 above). - Selected vehicle O* k the primary target (or absence of a primary target) identified in the previous sub-step 150; - Selected vehicle O* k Information only about, e.g. its lateral movement, and / or the on or off state of one of its indicators, and / or its stored location history, - Selected vehicle O* k Information about the current environment, e.g. Selected vehicle O* k the type of lane markings in which it is located (dotted or solid), and / or Selected vehicle O* kthe occupancy status of the lane adjacent to the current lane, and / or The current speed limit assigned to the lane (e.g. derived from a digital map pre-recorded in the GPS system or in-vehicle system, or extracted from images of traffic signs captured by an on-board camera on the target vehicle).
[0046] Selected vehicle O* k The maneuvers predicted for preferably form part of a predefined set of possible maneuvers, for example: -K.L. k , i.e., stay in its current lane: selected vehicle O* k continues driving in its current lane, -LLC k , i.e., lane change to the left lane: selected vehicle O* k will move to the adjacent lane to the left of the current lane. -RLC k , i.e. lane change to the right lane: selected vehicle O* k will move to the adjacent lane to the right of the current lane. -S k , i.e. stopped in the current lane: selected vehicle O* k will be stopped in its current lane.
[0047] The aforementioned maneuvers are not limiting: other maneuvers more suitable for other road configurations (e.g. reaching a roundabout or intersection) can be envisaged without departing from the scope of the invention.
[0048] Table 5 below presents, in particular, the primary targets, predicted maneuvers, current lanes, and target lanes obtained for various vehicles in the road scene shown in Figures 1 and 2 for prediction time t1 at the end of sub-step 160. [Table 5]
[0049] The in-vehicle system then detects the selected vehicle O* k In the coordinate system of the primary target (or the absence of a primary target) and the selected vehicle O* k Based on the estimated maneuver of the selected vehicle O* between the initial time t0 and the first predicted time t1, k The predicted movement may then be estimated in terms of the positions and velocities of the vehicles in the selected set (substep 170). Table 6 below provides an example of such predicted movement when the prediction process described above is performed in order for all vehicles in the selected set, according to priority. [Table 6]
[0050] The above results are then used to calculate the target vehicle O I t1, and then transformed (translated and rotated) again to be represented in the coordinate system linked to t2, so that, based on the calculated movements, it is possible to present a predicted position and velocity for a predicted time instant t1 of each vehicle (step 180, FIG. 3), which position and this velocity are related to the target vehicle O I For example, for track O3, the relative position {X;Y;θ} obtained at the end of the first prediction cycle relative to point P3(1) is 3,1 and speed
number
number
number
number
[0051] A second prediction cycle (not shown), similar to the first prediction cycle described above, estimates the points P occupied by each of the K vehicles in the set at a second prediction time t2, which is a time step Δt away from the first time t1. k To predict (2), i.e., in the first coordinate system and for a second prediction time, k The second relative position {X;Y;θ} k,2 and the second relative velocity
number
number
[0052] By generalizing the above principle, provision may be made to complete the method having a number N of consecutive prediction cycles greater than 2, each nth prediction cycle being calculated in a first coordinate system associated with the target vehicle and corresponding to a preceding prediction time instant t n-1 The nth prediction time point t n For each vehicle in the initial set, k nth relative position {X;Y;θ} k,n and the nth relative velocity
number
number
Claims
1. Target automobile vehicle (O I ) in an on-board driving assistance system on the target automobile vehicle (O I A driving assistance method (100) for a vehicle - the target automobile vehicle (O I The in-vehicle driving assistance system on the target automobile vehicle (O I ) in the environment of the vehicle (O 1 ~O 3 , O 5 ~O 7 ) in a detection step (110); - The target automobile vehicle (O I For each vehicle O in the set, for a first predicted time point following the initial time point in a first coordinate system associated with k a first prediction cycle of prediction of the first relative position and the first relative velocity of the first vehicle; the first prediction cycle comprising: The third party vehicle and the target automobile vehicle (O I ) for each vehicle O in a set of K vehicles k storing (120) initial data in a database of the in-vehicle driver assistance system, the initial data being a set of vehicle O in the first coordinate system at the initial time point; k (120), comprising information regarding the current lane, the current relative position, and the current relative speed of the vehicle; The K vehicles in the set k a sorting step (130) in which a priority is assigned to each vehicle O in the set stored in the database, k and the lane, and I ) in the multi-lane driving zone starting from the vehicle detected at the furthest forward position of k a sorting step (130) in which the Each vehicle in the set in the order of priority k A selection step (140) of selecting For each selected vehicle O* in the set k In contrast, - the selected vehicle O* k a first sub-step (150) of identifying another vehicle in said set that may be a primary target vehicle for - the identified primary target vehicle, the selected vehicle O* k and the selected vehicle O* k based at least on information about the current environment of the selected vehicle O* k a second sub-step (160) of estimating the maneuver in progress or to be executed by - the selected vehicle O* k Based on the estimated maneuver for the selected vehicle O* between the initial time point and the first predicted time point, k a third sub-step (170) of estimating the movement of Including, The selected vehicle O* for the first prediction time point k the first relative position and the first relative velocity of the vehicle are predicted based on the estimated movement; The first sub-step (150), the second sub-step (160) and the third sub-step (170) are performed to select the selected vehicle O* k based on position and velocity data transformed into a two-dimensional coordinate system associated with A driving assistance method (100).
2. The method according to claim 1, characterized in that the estimated maneuver is selected from a predefined set of possible maneuvers.
3. Selected vehicle O* k said predefined set of possible maneuvers for - the selected vehicle O* k maintain current lanes; - the selected vehicle O* k change lanes to the left, - the selected vehicle O* k change lanes to the right, - the selected vehicle O* k is the selected vehicle O* k Stop in your current lane, The method of claim 2, comprising:
4. The first substep (150) includes selecting the selected vehicle O* k 4. The method according to claim 1, further comprising searching for a vehicle in said set that is in front of and closest to the vehicle in the same lane.
5. The selected vehicle O* k The information relating only to the selected vehicle O* k and / or the lateral movement of the selected vehicle O* k and / or the on or off state of one of the indicators of the selected vehicle O*. k 5. The method according to claim 1, further comprising storing a history of the positions of the plurality of sensors.
6. The selected vehicle O* k The information about the current environment of the selected vehicle O* k the type of marking line of the lane in which the selected vehicle O* is located, and / or k 6. The method according to claim 1, further comprising determining whether the lane is currently occupied by a vehicle and / or whether the lane is currently occupied by a vehicle.
7. The method further comprises: calculating, by the in-vehicle driving assistance system, a time point in the first coordinate system for each vehicle O in the set for a second prediction time point subsequent to the first prediction time point. k The second relative position {X; Y; θ} k,2 and the second relative velocity [0010] a second prediction cycle of predictions of A first step of updating the database, in which initial relative position and relative velocity data are added to each vehicle O in the set k The first relative position {X;Y;θ} predicted in the first prediction cycle k,1 and the first relative velocity [0025] The first step is replaced by - the sorting step (130), the selection step (140), the first sub-step (150), the second sub-step (160) and the third sub-step (170) of the first prediction cycle, in which the initial time point has been replaced by the first prediction time point and the first prediction time point has been replaced by the second prediction time point; 7. The method according to claim 1, comprising:
8. The method includes a number N of consecutive prediction cycles, the number N being greater than 2, and each nth prediction cycle is a prediction cycle for each vehicle O in the set in the first coordinate system and for an nth prediction time instant following a preceding prediction time instant. k nth relative position {X; Y; θ} k,n and the nth relative velocity [0030] and each n-th prediction cycle is - updating said database, said relative position and relative velocity data being updated for each vehicle O in said set k with the relative position and relative velocity predicted in the preceding prediction cycle; and - the sorting step (130), the selection step (140), the first sub-step (150), the second sub-step (160) and the third sub-step (170) of the first prediction cycle in which the initial time point has been replaced by the preceding prediction time point and the first prediction time point has been replaced by the n-th prediction time point; The method of claim 7, comprising:
9. 9. The method of claim 8, wherein successive prediction time points are separated by a constant time step.
10. 10. The method of claim 9, wherein N is equal to 33 and the fixed time step is equal to 200 ms.
11. A target motor vehicle (O) carrying out the method according to any one of claims 1 to 10. I ) In-vehicle driving assistance system.
Citation Information
Patent Citations
Travel control device and travel control system
JP2010036862A
Driving support device
JP2015075889A
Vehicle control device, vehicle control method, and vehicle control program
WO2017010344A1
