Use of predefined (pre-built) graphical representations of roads for autonomous driving of vehicles and display of route planning

Abstract

The use of predefined (pre-built) graphical representations of roads for autonomous driving of vehicles and display of route planning.

The present invention relates to a method of generating and using a new data set for a novel technique of autonomous vehicle driving and route planning display using a predefined graphic representation/definition/illustration of the vehicle surroundings, in combination with the actual view sighted and captured by video cameras from the vehicle (for example, a dashboard camera). This method is primarily based on the use of the predefined graphic representation of the actual visual scene that is sighted from a host vehicle on roads at known GPS locations.

The main feature of this disclosure is the use of a predefined and simplified graphic illustration of the actual view experienced from vehicles on roads. This data is prepared in advance, stored in a cloud-based facility, and can be accessed by any vehicle.

The method comprises at least one video camera mounted in the vehicle, a module for accessing the predefined graphic visualization of the road and its surroundings, a module for defining the exact position and direction of the host vehicle on the predefined graphic view of the road, based on the use of GPS in combination with the identification of the surrounding characteristics of the road, a module continuously aligning the graphic view with the real view filmed by the video camera and for displaying the graphic visualization of the real view filmed by a forward facing camera.

The present invention relates to a new method for generating data that can be used in the vehicle's autopilot and for displaying route planning.

Numerous variations and modifications of the invention may be considered other than those described herein without departing from the spirit and scope of the present invention. This invention should not be limited by the specific details described herein.

Claims

1. Method of generating and using a new data set for a new technique of autonomous vehicle driving and route planning display, to be used on a vehicle equipped with one or more camera (s), one or more vehicle sensors and a communication interface connected to a cloud-based facility containing a simplified graphical representation of the visual road scene experienced from vehicles on the roads.

2. The method of claim 1 comprises the steps: Prepare in advance a database of pre-defined and simplified representation of visual scenes experienced from vehicles driving on roads; The pre-defined and simplified graphic representation of roads is a 360 degree graphical representation of the actual view of the road, as it can be experienced from a vehicle, using simple graphical and geometric features; The predefined and simplified graphic illustration of the roads has a coordinate system that corresponds to the GPS position on the roads; At each GPS location, the database includes information that identifies road limits, road markings, bridges, roundabouts, and other road information; The predefined graphic representation does not include moving or stationary objects (buildings, cars, bicycles, etc.). It does not include any other object not directly related to roads geometrical information; The predefined database includes representations of roads and surroundings (traffic lanes, road markings, traffic signs, roundabouts, road bridges, etc.) and other road information in an empty space or empty background; The database covers a network of roads. The database is stored in a cloud-based facility for later access by any vehicle. It consists of a simplified graphic/geometric representation of the actual view of roads as sighted from vehicles on roads and it is associated with a GPS location; Data collected by a GPS system on board the vehicle; Data collected with the vehicle's camera (s), including forwarding facing camera; Data collected with the vehicle's sensor (5); Continuous and real-time access to the pre-defined and simplified 360 degree representation of the view from the vehicle at each GPS location; Continuously adjust the visual scene recorded by the camera (s) with the simplified 360 degree representation at the GPS location of the vehicle, and continuously adjust the vehicle position to the exact and corresponding position and direction in the predefined and simplified graphic representation database.

3. The method of claim 1, wherein preparing in advance a predefined and simplified graphic representation of the 360 degree visual scenes experienced from a vehicle on the road further comprises using a special road view recording vehicle equipped with cameras and a geolocation unit establishing the geographical position of the special road view recording vehicle and associating it to the filmed view of the continuously recorded scene.

4. A method according to claim 1, wherein the special road view recording vehicle, comprises cameras at different locations and at different orientations to collect in advance the 360 degree view at any location on the road and at different heights of the road surface.

5. The method of claim 1, wherein preparing in advance a predefined simplified graphic representation of the 360 degree visual scenes experienced from a vehicle on the road further comprises collecting, in advance, video data under the form a 360 degree video footage and/or images of the actual scene filmed by one or more cameras installed in a special road view recording vehicle. Data is collected for a road networks.

6. The method of claim 1, wherein preparing in advance a predefined simplified graphic representation of the 360 degree visual scenes from a vehicle on the road further comprises converting the filmed view into a simplified illustration of the view filmed in advance by the specially equipped vehicle. This simplified illustration consists of simple geometric and graphic elements such as lines and simple shapes to represent road limits, traffic lanes, road marking and traffic signs. Geometric entities can be represented/displayed in empty space or empty background. The predefined simplified graphic representation database is stored in a cloud-based facility for later access by any vehicle.

7. The method of claim 1 wherein a host vehicle comprising a communication unit continuously receiving the predefined and simplified graphic representation of the 360 degree visual scenes from a vehicle on the road at the GPS location of the vehicle

8. The method of claim 1 wherein the data collected with the host vehicle camera (s) further comprises collecting video footage of the actual scene in front of the vehicle filmed by a forward-facing camera. Additional cameras mounted on the vehicle and pointing in different directions gather video footage of the actual scene around the vehicle.

9. The method of claim 1 further comprising processing the video footage, identifying reference features, such as road limits, road marking and road lanes, and matching the exact location and direction of the host vehicle in the predefined 360 degree graphic representation database.

10. The method of claim 1 further comprising collecting data from external sensors in the form of a relative location, relative speed, and relative acceleration between the vehicle and another vehicle or objects.

11. The method of claim 1 further comprising identifying the location of other vehicles and/or objects in the predefined 360 degree database.

12. The method of claim 1 further comprising converting the information gathered from the predefined 360 degree visualization into autonomous travel instructions.

13. The method of claim 1, further comprising a display unit mounted in the host vehicle for displaying in real time the adjusted pre-defined representation of the view recorded by the forward facing camera, and which include an illustration of the identified vehicles and objects in their exact location.

Description

BRIEF DESCRIPTION OF THE SEVERAL VIEWS OF THE DRAWING

[0022] The present disclosure is hereinafter described in detail in connection with the accompanying figures. The figures are used by way of illustration only and provide exemplary embodiments.

[0023] FIG. 1 is an illustration of the special vehicle equipped with cameras and a storage unit, which is used to record the data required for the definition of the pre-defined 360 degree graphic illustration of the view of roads as experienced from vehicles on roads

[0024] FIG. 2 illustrates a vehicle equipped with the instruments required to operate the method in this disclosure.

[0025] FIG. 3 Is a flowchart depicting the steps for the embodiment of this disclosure

[0026] FIG. 4 is an illustrative example of the 360 degree database generation steps at a GPS location on a road.

[0027] FIG. 5 is an illustrative example of the view filmed by a forward-facing camera on a given vehicle and the data extracted from the predefined 360 degree graphic database at the vehicle's GPS location corresponding to the view filmed by the camera facing forward.

DETAILED DESCRIPTION OF THE INVENTION

[0028] The method described herein is designed to support developing a new technique for vehicle autonomous driving and route planning display, based on the use of the predefined and simplified graphic illustrations of the roads and their surroundings, in conjunction with the real-time location of the vehicle and vehicle external data, which is collected using vehicle mounted sensors.

[0029] In accordance with various embodiments, the pre-defined simplified graphic representation is prepared in advance from the collection and processing of visual data at various locations on the roads. Visual data corresponds to the 360 degree landscape as experienced from any possible location of a vehicle on the road. The predefined and simplified graphic database is called later to identify the precise location and direction of a given vehicle in the predefined database and to be used as part of a new autonomous driving method.

[0030] The 360 degree graphic and simplified representation of the visual seen from a vehicle is prepared in advance using a specially equipped vehicle with cameras to film the view in the road at 360 degree angle and different position on the road and height from the surface of the road. The filmed visual scene is processed and simplified to keep only a simplified graphic geometric representation of: the road boundaries, road limits, road lanes, road marking and a geometric illustration of some standard features like bridges and roundabout, in an empty background. The simplified 360 degree graphic representation is stored in a cloud-based database. Each location on the simplified 360 degree graphic representation is linked to a GPS coordinate and precise location on the road.

[0031] In FIG. 1, there is shown an example of a special road view recording vehicle 100 equipped with video cameras 110, a data storage unit 120 and a navigation unit or a GPS 130. The special vehicle 100 is used to collect the necessary information for the preparation of the simple and pre-defined graphic representation of the visual scenes experienced from vehicles on roads. The vehicle 100 drives on the roads to film the visual scene captured by these video cameras 110 and to store the corresponding data in the data storage unit 120. The cameras 110 are pointed in different directions and mounted at different heights from the road surface. The objective is to collect 360 degree views at different heights and be able to reconstruct the 360 degree view at any other possible height.

[0032] The data storage unit 120 is used to collect and store the information recorded by the cameras mounted on the special vehicle 100. A navigation unit or a GPS 130 continuously establishes the geographical location of the vehicle and links it to the data recorded by the video cameras 110 to associate a GPS position with the recorded video views.

[0033] FIG. 2 shows an example of a host vehicle 200 having a device 210 including an autonomous driving unit and a route planning display. In this example, the vehicle device 210 is equipped with a hardware component 220 which includes multiple cameras 230, a GPS unit 240, a data processing module 250, one or more external sensor (s) 260, a communication unit 270, an autonomous vehicle driving unit 280, a display unit 290, other suitable electronic components as well as any other hardware, software, etc. which are needed to implement the method described herein. The vehicle hardware 220 arrangement of FIG. 2 is for illustration only.

[0034] Vehicle video cameras 230 are mounted on vehicle 200, at different locations and pointed in different directions for 360 degree visual coverage of the surrounding vehicle scene. One of the cameras faces forward to capture the visual scene in front of the vehicle.

External sensors 260 located around the vehicle may include, but are not limited to, a radar unit and/or an ultrasound unit. The external vehicle sensor provides information to system 210 regarding the relative position of objects and/or vehicles on the road from vehicle 200.

[0035] The autonomous vehicle driving module 280 provides the vehicle with autonomous driving capabilities. Module 280 includes a component for communicating with system 250 to collect the exact and corresponding location of vehicle 200 in the 360 degree graphic database, and a route planning software for calculating a travel route. The predefined graphical representation of the view recorded by the forward-facing camera of cameras 220, which is retrieved from the predefined database, can be viewed using display unit 290.

[0036] FIG. 3 shows a flowchart which illustrates the different stages of implementation of this method from the generation and storage of the predefined 360 degree graphic view of the visual scene experienced from vehicles on roads, to the operations that can be carried out according to various embodiments to generate data set for autonomous driving and display of route planning. From step 300 where a database is once created and is available in a cloud-based system 310 and accessible from any vehicle 200 equipped with instruments 210, 220, 230, 240, 250, 260, 280 and 290. The geolocation of the vehicle 200 is tracked and used to identify in real time the location of the vehicle 200 in the cloud-based database 310, which consists of a 360 degree simplified graphic representation. Step 320 may use multiple video cameras 230 mounted on the vehicle 200, located around the vehicle 200 and facing different directions, including the forward direction looking at an area in front of the vehicle 200. The video data of the vehicle are collected and provided in any data format to the data processing module 250 and/or some other components. The data recorded by the vehicle's cameras is used by a data processing module 250, and/or another component, to refine the location of the vehicle in the 360 degree database by identifying road features, such as road boundaries and road markings, and continuously match the exact location and direction of the vehicle in the pre-defined 360 degree database. Step 320 uses vehicle sensors consisting of radar and/or an ultrasound-based system 260 to collect external data from the vehicle and provide it to the vehicle data processing module and/or some other components 250. Vehicle sensors identify vehicles and objects around vehicle 200 and their distance and relative location from vehicle 200. Vehicle sensor data is processed by a data processing module 250 to identify the exact location of the surrounding vehicles and objects in the predefined 360 degree database 310. The processed data, consisting of the exact location of the vehicle 200 on the predefined 360 degree database and the location of the external objects and other vehicles on the road is communicated to an autonomous driving unit 280 to translate this information into a driving action or function, without driver request, to support autonomous driving.

[0037] FIG. 4 illustrates the operations, stage 400 and stage 410, that can be performed according to various embodiments to generate the predefined 360 degree graphic illustration of the visual data recorded by the different cameras cited in FIG. 1. The video cameras 110 are mounted on the vehicle 100 at different height from the road surface and located around the vehicle and facing different directions, including the forward direction. The video cameras 110 are intended to collect video images at 360 degrees and at different heights of the environment around the vehicle 100. The data obtained from the recording, stage 400, using the special vehicle 100 is converted in stage 410 into a simplified graphic illustration of the recorded view using computer graphics technology. The predefined and simplified 360 degree graphic illustration includes simple drawings of the road, road limits, road marking, traffic lanes and specific features such as roundabouts and bridges. As shown in FIG. 4, geometric lines are used in an empty background to represent the characteristics of the road. Objects on the road and its surrounding are not represented in the simplified 360 degree graphic definition. Only the continuous representation of roads and special traffic elements such as road marking, bridges, traffic signs and traffic lights are displayed. FIG. 4 also shows, by way of illustration only, the view at a given GPS position which can be filmed by the cameras mounted on the vehicle 100, facing forward, positioned at the same height from the road surface, but at different locations relative to the transverse direction of the road or width of the road. The views recorded by the other cameras of cameras 110 of the vehicle 100 are not shown in FIG. 4. The processing of data from video cameras 110 and GPS 130 and covering a large network of roads, using computer graphics technology, generates a 360 degree simplified graphic illustration of the road as shown as an example in FIG. 4. The predefined 360 degree representation of the road network is stored in a cloud-based database.

[0038] In FIG. 5, the data collected by the video camera 230 and the GPS location from the GPS unit 240 are provided in any data format to the vehicle data processing module and/or some other system components 250. The GPS location of the vehicle 200 is also provided continuously to the communication unit 270, which then connects to the predefined cloud-based 360 degree database to extract the graphic and simplified representation of the road view at that location. All the information collected by components 230, 240 and 270 are used by the data processing module and/or some other system components 250 to identify reference features, such as, but not limited to, road boundaries, road markings and road lanes and to continuously identify the exact position and direction of the vehicle 200 in the 360 degree pre-defined graphics database that matches the exact and actual location and direction of the vehicle on the road captured by video cameras 230. The component 250 can correlate the images from the camera to the exact location of the vehicle in the predefined graphic representation database.

[0039] Also in FIG. 5 there is shown an illustration of how the information extracted from the pre-defined 360 degree graphic database is matched by component 250 with the actual view filmed by the forward-facing camera of cameras 230. The display device 290 can provide in real-time the view extracted from the pre-defined database, that correlates to and matches the view recorded by the forwarding facing camera.

[0040] The term vehicle designates a road transport machine such as a car.

[0041] The term host vehicle is used to refer to any vehicle on the road that is enabled to use the present invention for autonomous driving of vehicles and display of route planning.

[0042] The term special road view recording vehicle is used to refer to a vehicle specially equipped with the appropriate cameras, instruments and programming software to record continuously the vehicle surrounding view while the vehicle is travelling.