Self-adaptive pedestrian protection system and method based on automatic driving park vehicle

Abstract

The invention discloses a self-adaptive pedestrian protection system and method based on an automatic driving park vehicle, and belongs to the technical field of automobile automatic driving. The self-adaptive pedestrian protection system comprises an automatic driving real vehicle control module, a camera sensing module, an ultrasonic sensing module, a pedestrian protection control module and a self-adaptive pedestrian protection execution module. According to the self-adaptive pedestrian protection method, the self-adaptive pedestrian protection system comprehensively analyzes pedestrian detection data based on a camera and an ultrasonic radar, road edge obstacle information, vehicle speed and other signals, and actively controls the anti-collision flexible strips arranged on the periphery of a vehicle body to expand outwards and contract inwards so as to prevent pedestrians from colliding with the vehicle body. And meanwhile, after the pedestrian inevitably collides with the anti-collision flexible strip, the flexible strip is compressed to enable a switching circuit pre-embedded in the flexible strip to reach a conduction condition, so that emergency braking of the vehicle is triggered to avoid dangerous collision working conditions, and the effect of protecting the pedestrian to the greatest extent is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of automotive autonomous driving, and more specifically, to an adaptive pedestrian protection system and method based on an autonomous driving park vehicle. Background Art

[0002] With the continuous increase in the commercial penetration rate of autonomous driving park vehicles (including cleaning vehicles, express delivery vehicles, security vehicles, etc.), more and more related vehicles are driving in some semi-closed parks. Although the road planning in semi-closed parks is relatively fixed, the mixing of people and vehicles is frequent (such as the coexistence of electric assist vehicles, patrol vehicles, employees, visitors, etc.); although the vehicle speed is relatively low (<30 km / h), there are still collision risks in scenarios such as sudden stops and turns; pedestrians may suddenly cross, stay, or carry equipment (such as handcarts), which requires extremely high real-time response capabilities of the autonomous driving system; the cameras and lidars of autonomous driving vehicles may be blocked by obstacles (such as cargo stacks, green belts), resulting in sensor blind spots; in extreme weather (rain, snow, fog) or backlight conditions, the accuracy of pedestrian recognition decreases, and the algorithm has certain limitations. All of the above factors may increase the danger to pedestrians during vehicle operation. Therefore, it is crucial to equip autonomous driving park vehicles with advanced and reliable pedestrian protection systems.

[0003] Currently, the equipment rate of pedestrian protection devices for park vehicles is already very high, and most vehicles are equipped with protection devices such as passive anti-collision strips that can trigger emergency braking of the vehicle. For example, the Chinese utility model patent with the publication number CN 212579782U discloses a pedestrian protection device for a driverless vehicle. This pedestrian protection device for a driverless vehicle protects pedestrians by setting a buffer belt and a baffle, and although it has the advantages of simple structure, convenient maintenance, protecting people's legs and preventing wheel rolling, it cannot actively adapt its protection measures according to the pedestrian flow density, pedestrian movement speed, etc., and cannot further improve the protection effect. Summary of the Invention

[0004] In view of the above-mentioned defects (deficiencies) of the prior art, the purpose of the present invention is to provide an adaptive pedestrian protection system and method based on an autonomous driving park vehicle.

[0005] The present invention adopts the following technical solutions: An adaptive pedestrian protection system based on an autonomous driving park vehicle, comprising: an autonomous driving real vehicle control module, a camera sensing module, an ultrasonic sensing module, a pedestrian protection control module, and an adaptive pedestrian protection execution module; The autonomous driving real vehicle control module, which has the control system of a park vehicle with autonomous driving function, is the basic carrier for the normal operation of other modules; The camera perception module includes four cameras, which are distributed and installed on the edge of the vehicle roof, and respectively monitor the pedestrian distribution and movement status in the front, rear, left and right areas of the vehicle, and give pedestrian data information including coordinate position, movement speed and whether approaching the vehicle for each pedestrian target; The ultrasonic perception module includes twelve ultrasonic probes, with two installed on the front and rear enclosures of the vehicle respectively, and four installed on the left and right enclosures respectively, providing precise perception of obstacles within 3 meters around the vehicle body; The obstacles include moving vehicles, parked vehicles, and raised road curbs, etc.

[0006] The pedestrian protection control module simultaneously monitors vehicle signals including vehicle speed, pedestrian data information sent by the camera perception module, and obstacle detection information sent by the ultrasonic perception module. After comprehensive judgment, it controls the anti-collision flexible strips of the adaptive pedestrian protection execution module to perform expansion and contraction actions, enabling the vehicle safety frame to have the ability of dynamic change; The adaptive pedestrian protection execution module includes four anti-collision strips and four telescopic devices. The four anti-collision strips are respectively arranged and installed on the bottom sides of the front, rear, left and right of the vehicle body. Each anti-collision strip is connected and fixed to the vehicle body through one of the telescopic devices to individually control the expansion and contraction of each anti-collision strip; The adjacent two anti-collision strips are connected by a telescopic material; The outer surface of each anti-collision strip is coated with a flexible material, and a parallel plate switch circuit is built into the flexible material. When deformed by external force extrusion, the parallel plate switch circuit is turned on to transmit the collision extrusion signal to the pedestrian protection control module.

[0007] The above control system controls the park vehicle to achieve basic autonomous driving functions including steer-by-wire, drive-by-wire, brake-by-wire, centimeter-level positioning, following the track along the built-in map, and lane-changing to avoid obstacles.

[0008] The present invention also provides an adaptive pedestrian protection method for an autonomous driving park vehicle, using the above-mentioned adaptive pedestrian protection system for an autonomous driving park vehicle, including the following steps: (1) Start the software and hardware modules related to the adaptive pedestrian protection system of the autonomous driving park vehicle; (2) Judge whether there are key faults affecting the system operation. If not, enter step (3); (3) The autonomous driving park vehicle automatically drives according to the preset test scenario; (4) Collect data from the camera perception module and the ultrasonic perception module; (5) The pedestrian protection control module processes the data from the camera perception module and the ultrasonic perception module, and sends control instructions to the adaptive pedestrian protection execution module; (6) The adaptive pedestrian protection execution module executes the instructions, judges whether there is an anti-collision strip collision signal. If so, enter step (7); if not, return to step (4); (7) Sending an emergency braking request to the autonomous driving vehicle control module; (8) Safe parking of autonomous vehicles in the park.

[0009] The specific strategy of the above step (5) is as follows: S51, real-time data retrieval and extraction, determine whether it is valid data, if so, proceed to step S52; S52, determine whether there are pedestrians within a radius of 6 meters from the center point of the vehicle, if so, proceed to step S53; S53, determine whether there are pedestrians within a radius of 3 meters from the center of the vehicle; if yes, proceed to step S54; if no, proceed to step S57; S54, determine whether there are pedestrians within a radius of 1 meter from the center of the vehicle; if yes, proceed to step S55, if not, proceed to step S56; S55. Calculate the outward expansion distance of the anti-collision flexible strip: D=Dmax, where Dmax indicates the maximum supported outward expansion distance filled in first; S56. Calculate the outward expansion distance of the anti-collision flexible strip: D = N*V*Dmax / 36*120%, where Dmax indicates the maximum supported outward expansion distance; N indicates the number of people. If the number is greater than 10, it is calculated as 10 people; V indicates the vehicle speed in meters per second. If the number is greater than 3.6, it is calculated as 3.6; S57. Calculate the outward expansion distance of the anti-collision flexible strip: D = N*V*Dmax / 36, where Dmax indicates the maximum supported outward expansion distance; N indicates the number of people. If the number is greater than 10, it is calculated as 10 people; V indicates the vehicle speed in meters per second. If the number is greater than 3.6, it is calculated as 3.6; S58, determining whether the vehicle is in parking mode, if so, executing the restriction condition D=0, otherwise skipping this step; S59, determine whether the distance Dy between the curb and the vehicle body is less than D+0.1 meters. If so, execute D=Dy-0.1 meters. If not, perform adaptive pedestrian protection operation.

[0010] In the above step S58, being in parking mode means that the vehicle speed remains at 0 for more than 5 seconds.

[0011] It can be seen from the above description of the present invention that, compared with the prior art, the present invention has the following advantages: The adaptive pedestrian protection system of the present invention comprehensively analyzes pedestrian detection data based on cameras and ultrasonic radars, roadside obstacle information, vehicle speed and other signals, and actively controls the anti-collision flexible strips arranged around the vehicle body to expand and retract inward to prevent collision between pedestrians and the vehicle body. At the same time, when pedestrians inevitably collide with the anti-collision flexible strips, the flexible strips will be compressed so that the switch circuits embedded inside them reach the conduction condition, thereby triggering the vehicle's emergency braking to avoid dangerous collision conditions, thereby protecting pedestrians to the greatest extent. Description of the Drawings

[0012] Figure 1 This is the system module block diagram of the present invention.

[0013] Figure 2 This is the top view of the expanded state of the adaptive pedestrian protection execution module of the present invention.

[0014] Figure 3 This is the top view of the contracted state of the adaptive pedestrian protection execution module of the present invention.

[0015] Figure 4 This is the method flow chart of the system of the present invention.

[0016] Figure 5 This is the flow chart of the adaptive pedestrian protection of the present invention. Detailed Description of the Invention

[0017] The following will refer to Figure 1 to describe the detailed implementation of the present invention. To fully understand the present invention, many details are described below. However, for those skilled in the art, the present invention can be implemented without these details. For well-known components, methods, and processes, no further detailed description will be given below.

[0018] An adaptive pedestrian protection system based on an autonomous driving park vehicle, referring to Figure 1 , includes: an autonomous driving real vehicle control module, a camera perception module, an ultrasonic perception module, a pedestrian protection control module, and an adaptive pedestrian protection execution module.

[0019] The autonomous driving real vehicle control module has the control system of the autonomous driving park vehicle and is the basic carrier for the normal operation of other modules. The control system controls the park vehicle to achieve basic autonomous driving functions including steer-by-wire, drive-by-wire, brake-by-wire, centimeter-level positioning, following the built-in map, and lane-changing around obstacles.

[0020] The camera perception module includes four cameras, which are distributed and installed on the edge of the vehicle roof to monitor the pedestrian distribution and movement status in the front, rear, left, and right areas of the vehicle respectively, and give pedestrian data information including coordinate position, movement speed, and whether approaching the vehicle for each pedestrian target. The coordinates take the center point of the rear axle of the vehicle as the origin, the forward direction of the vehicle as the positive x-axis direction, the left side of the vehicle as the positive y-axis direction, and the vertical upward direction as the positive z-axis direction.

[0021] The ultrasonic perception module includes twelve ultrasonic probes, with two installed on each of the front and rear enclosures of the vehicle and four installed on each of the left and right enclosures, providing precise perception of obstacles within 3 meters around the vehicle body. The obstacles include moving vehicles, other vehicles, and raised road curbs, etc.

[0022] The pedestrian protection control module simultaneously monitors vehicle signals including vehicle speed, pedestrian data information sent by the camera perception module, and obstacle detection information sent by the ultrasonic perception module. After comprehensive judgment, it controls the anti-collision flexible strips of the adaptive pedestrian protection execution module to expand and contract, enabling the vehicle safety frame to have the ability of dynamic change and improving the pedestrian protection efficiency.

[0023] Refer to Figure 2 and Figure 3 . Figure 2 and Figure 3 In, 1 represents the stretchable material for connecting the anti-collision strip, 2 represents the telescopic device, 3 represents the vehicle body, and 4 represents the anti-collision strip. The adaptive pedestrian protection execution module includes four anti-collision strips 4 and four telescopic devices 2. The four anti-collision strips 4 are respectively arranged and installed on the bottom sides of the front, rear, left, and right of the vehicle body 3. Each anti-collision strip 4 is connected and fixed to the vehicle body 3 through a telescopic device 2 to independently control the expansion and contraction of each anti-collision strip 4; the adjacent two anti-collision strips 4 are connected by a stretchable material 1; the outer surface of each anti-collision strip 4 is coated with a flexible material, and a parallel plate switch circuit is built into the flexible material. The parallel plate switch circuit is not conductive in the normal state and is conductive when deformed by external force extrusion to transmit the collision extrusion signal to the pedestrian protection control module.

[0024] Refer to Figure 3 , the above-mentioned adaptive pedestrian protection system based on an autonomous driving campus vehicle, its adaptive pedestrian protection method includes the following steps: (1) Start the software and hardware modules related to the adaptive pedestrian protection system of the autonomous driving campus vehicle; (2) Determine whether there are key faults affecting the system operation. If not, enter step (3); (3) The autonomous driving campus vehicle automatically drives according to the preset test scenario; (4) Collect data from the camera perception module and the ultrasonic perception module; (5) The pedestrian protection control module processes the data from the camera perception module and the ultrasonic perception module and sends control instructions to the adaptive pedestrian protection execution module; (6) The adaptive pedestrian protection execution module executes the instructions, determines whether there is an anti-collision strip collision signal. If so, enter step (7); if not, return to step (4); (7) Send an emergency braking request to the autonomous driving real vehicle control module; (8) The autonomous driving campus vehicle safely stops.

[0025] As Figure 4 shown, the specific strategy of the above step (5) is as follows: S51. Retrieve and extract real-time data, and determine whether it is valid data. If so, enter step S52.

[0026] S52. Determine whether there are pedestrians within a 6-meter radius from the center point of the vehicle. If so, proceed to step S53.

[0027] S53, determine whether there are pedestrians within a radius of 3 meters from the center of the vehicle; if yes, proceed to step S54; if not, proceed to S57.

[0028] S54. Determine whether there are pedestrians within a 1-meter radius from the center of the vehicle; if so, proceed to step S55; if not, proceed to S56.

[0029] S55. Calculate the outward expansion distance of the anti-collision flexible strip: D=Dmax, where Dmax represents the maximum supported outward expansion distance.

[0030] S56. Calculate the outward expansion distance of the anti-collision flexible strip: D = N*V*Dmax / 36*120%, where Dmax indicates the maximum supported outward expansion distance; N indicates the number of people. If the number is greater than 10, it is counted as 10 people; V indicates the vehicle speed in meters per second. If the number is greater than 3.6, it is counted as 3.6.

[0031] S57. Calculate the outward expansion distance of the anti-collision flexible strip: D = N*V*Dmax / 36, where Dmax indicates the maximum supported outward expansion distance; N indicates the number of people. If it is greater than 10, it is counted as 10 people; V indicates the vehicle speed in meters per second. If it is greater than 3.6, it is counted as 3.6.

[0032] S58, determine whether the vehicle is in parking mode, if so, execute the restriction condition D=0, otherwise jump out of this step.

[0033] S59, determine whether the distance Dy between the curb and the vehicle body is less than D+0.1 meters. If so, execute D=Dy-0.1 meters. If not, perform adaptive pedestrian protection operation.

[0034] A typical application of the system of the present invention is given below. The purpose of this embodiment is to demonstrate the system function implementation method, data processing method and protection function execution status through the description of the system adaptive pedestrian protection function process in a typical scenario. The specific steps are as follows: Step 1 Record a GPS (Global Positioning System) map for the autonomous driving vehicle in a closed test site. The total length of the map should be more than 400 meters. It is recommended to make it into a small circle loop mode. Set up several dummies at appropriate locations on the road (some dummies need to be movable). Debug the autonomous driving vehicle so that it has the ability to automatically follow the GPS map and automatically perform emergency braking.

[0035] Step 2 Collect data from the camera perception module and the ultrasonic perception module; the data collected from the camera perception module shall at least include the number of pedestrian targets in the detection area and the position coordinates of each pedestrian target relative to the vehicle body; the data collected from the ultrasonic perception module shall at least include the number of obstacle targets in the detection area, the position coordinates of each obstacle target relative to the vehicle body, whether the obstacle is moving, and the distance between the curb (the convex type that can be detected) and the vehicle edge.

[0036] Process the collected targets, and the processing measures include: if the continuous duration of effective target detection is less than 400 milliseconds, then determine this target as an invalid target; for the obstacle target data detected by the ultrasonic perception module, the target within 3 meters from the center point of the vehicle and moving is regarded as an effective target; the accumulated values of the target data of the camera perception module and the ultrasonic perception module after screening are used as the basis for the number of effective targets determined in the subsequent steps.

[0037] Step 3 Run the adaptive pedestrian protection program: send corresponding control instructions to the adaptive pedestrian protection actuator based on the data of the camera and ultrasonic sensors processed in Step 2.

[0038] The specific adaptive pedestrian protection strategy is as follows: (1) If there is a pedestrian target within a radius of 1 meter (inclusive) from the center point of the vehicle, calculate the outward expansion distance of the anti-collision flexible strip: D = Dmax, where Dmax (unit: meter) represents the maximum supportable outward expansion distance filled in first.

[0039] (2) If there is a pedestrian target outside a radius of 1 meter (exclusive) and within 3 meters (inclusive) from the center point of the vehicle, calculate the outward expansion distance of the anti-collision flexible strip: D = N * V * Dmax / 36 * 120%, where Dmax represents the maximum supportable outward expansion distance filled in first, N represents the number of people (counted as 10 when greater than 10), and V represents the vehicle speed value (unit: meter per second, counted as 3.6 when greater than 3.6).

[0040] (3) If there is a pedestrian target outside a radius of 3 meters (exclusive) and within 6 meters (inclusive) from the center point of the vehicle, calculate the outward expansion distance of the anti-collision flexible strip: D = N * V * Dmax / 36, where Dmax represents the maximum supportable outward expansion distance filled in first, N represents the number of people (counted as 10 when greater than 10), and V represents the vehicle speed value (unit: meter per second, counted as 3.6 when greater than 3.6).

[0041] (4) Determine whether the vehicle is in the parking mode (the vehicle speed remains 0 for more than 5 seconds). If so, execute the limiting condition D = 0, otherwise skip this step.

[0042] (5) Determine whether the distance Dy between the curb and the vehicle body is less than D + 0.1 meter. If so, execute D = Dy - 0.1m.

[0043] (6)The curb-to-vehicle distance Dy determined in (5) is controlled separately for the front, rear, left, and right sides, that is, the calculated D values are D_Q / D_H / D_Z / D_Y for the front, rear, left, and right sides of the vehicle respectively. If a certain side meets the determination condition described in (5), the D value control strategy is executed respectively to calculate the final control amount.

[0044] The calculated D_Q / D_H / D_Z / D_Y control amounts are sent to the pedestrian protection actuator to separately control the telescopic amounts of the four anti-collision strips on the front, rear, left, and right sides of the vehicle body.

[0045] Step 4 Monitor whether there is a collision signal on the anti-collision flexible strip. If there is, send an emergency braking request to the autonomous driving vehicle control module to stop the vehicle in time to maximize the safety of other traffic participants.

[0046] Step 5 Loop through steps 2 - 5 to enable the autonomous driving vehicle to continuously and dynamically execute the active adaptive pedestrian protection function throughout the entire operation process, effectively improving driving safety.

[0047] The above is only the specific implementation manner of the present invention, but the design concept of the present invention is not limited thereto. Any non-substantive modification made to the present invention using this concept shall fall within the scope of infringement of the protection scope of the present invention.

Claims

1. An adaptive pedestrian protection system based on an autonomous campus vehicle, characterized in that, Including: An autonomous driving vehicle control module, a camera perception module, an ultrasonic perception module, a pedestrian protection control module, and an adaptive pedestrian protection execution module; The autonomous driving vehicle control module, which has the control system of the autonomous driving function park vehicle, is the basic carrier for the normal operation of other modules; The camera perception module includes four cameras, which are distributed and installed on the edge of the vehicle roof, and respectively monitor the pedestrian distribution and movement status in the front, rear, left, and right areas of the vehicle, and give pedestrian data information including coordinate position, movement speed, and whether the pedestrian is approaching the vehicle for each pedestrian target; The ultrasonic perception module includes twelve ultrasonic probes, with two installed on the front and rear enclosures of the vehicle respectively, and four installed on the left and right enclosures respectively, providing precise perception of obstacles within 3 meters around the vehicle body; The pedestrian protection control module simultaneously monitors vehicle signals including vehicle speed, pedestrian data information sent by the camera perception module, and obstacle detection information sent by the ultrasonic perception module. After comprehensive judgment, it controls the anti-collision flexible strips of the adaptive pedestrian protection execution module to expand and contract, enabling the vehicle safety frame to have the ability to dynamically change; The adaptive pedestrian protection execution module includes four anti-collision strips and four telescopic devices. The four anti-collision strips are respectively arranged and installed on the bottom sides of the front, rear, left, and right of the vehicle body, and each anti-collision strip is connected and fixed to the vehicle body through one of the telescopic devices to control the outward expansion and contraction of each anti-collision strip independently; The outer surface of each anti-collision strip is coated with a flexible material, and a parallel plate switch circuit is built into the flexible material. When it is deformed by external force extrusion, the parallel plate switch circuit is turned on to transmit the collision extrusion signal to the pedestrian protection control module.

2. The adaptive pedestrian protection system based on an autonomous driving campus vehicle according to claim 1, characterized in that: The control system controls the park vehicle to realize basic autonomous driving functions including steer-by-wire, drive-by-wire, brake-by-wire, centimeter-level positioning, following the track along the built-in map, and lane-changing and obstacle avoidance.

3. An adaptive pedestrian protection system based on an autonomous driving park vehicle according to claim 1, characterized in that: The obstacles include moving vehicles, parked vehicles, and raised road curbs.

4. An adaptive pedestrian protection system based on an autonomous driving park vehicle according to claim 1, characterized in that: The adjacent two anti-collision strips are connected by a telescopic material.

5. An adaptive pedestrian protection method based on an autonomous driving park vehicle, which adopts an adaptive pedestrian protection system based on an autonomous driving park vehicle as described in claim 1, characterized in that, Including the following steps: (1) Start the software and hardware modules related to the adaptive pedestrian protection system of the autonomous driving park vehicle; (2) Judge whether there is a key fault affecting the system operation. If not, enter step (3); (3) The autonomous driving park vehicle automatically drives according to the preset test scenario; (4) Collect data from the camera perception module and the ultrasonic perception module; (5) The pedestrian protection control module processes the data from the camera perception module and the ultrasonic perception module, and sends control instructions to the adaptive pedestrian protection execution module; (6) The adaptive pedestrian protection execution module executes the instructions, judges whether there is an anti-collision strip collision signal. If so, enter step (7); if not, return to step (4); (7) Send an emergency braking request to the autonomous driving vehicle control module; (8) The autonomous driving park vehicle stops safely.

6. The adaptive pedestrian protection method based on an autonomous driving park vehicle according to claim 5, wherein: If there is a key fault affecting the system operation in step (2), then judge whether the fault can be eliminated by a simple restart. If so, enter step (1); if not, the autonomous driving park vehicle ends its operation.

7. The adaptive pedestrian protection method based on an autonomous driving campus vehicle according to claim 1, characterized in that, The specific strategy of step (5) is as follows: S51. Retrieve and extract real-time data, and determine whether it is valid data. If so, proceed to step S52; S52. Determine whether there are pedestrians within a radius of 6 meters from the vehicle center point. If so, proceed to step S53; S53. Determine whether there are pedestrians within a radius of 3 meters from the vehicle center. If so, proceed to step S54; if not, proceed to S57; S54. Determine whether there are pedestrians within a radius of 1 meter from the vehicle center. If so, proceed to step S55; if not, proceed to S56; S55. Calculate the outward expansion distance of the anti-collision flexible strip: D = Dmax, where Dmax represents the maximum supportable outward expansion distance filled in first; S56. Calculate the outward expansion distance of the anti-collision flexible strip: D = N * V * Dmax / 36 * 120%, where Dmax represents the maximum supportable outward expansion distance filled in first; N represents the number of people, counted as 10 when greater than 10; V represents the vehicle speed value in meters per second, counted as 3.6 when greater than 3.6; S57. Calculate the outward expansion distance of the anti-collision flexible strip: D = N * V * Dmax / 36, where Dmax represents the maximum supportable outward expansion distance filled in first; N represents the number of people, counted as 10 when greater than 10; V represents the vehicle speed value in meters per second, counted as 3.6 when greater than 3.6; S58. Determine whether the vehicle is in the parking mode. If so, execute the limiting condition D = 0; otherwise, jump out of this step; S59. Determine whether the distance Dy between the road edge and the vehicle body is less than D + 0.1 meters. If so, execute D = Dy - 0.1m; if not, execute the adaptive pedestrian protection operation.

8. An adaptive pedestrian protection method based on an autonomous driving park vehicle according to claim 1, characterized in that: In step S58, being in the parking mode means that the vehicle speed remains 0 for a duration greater than 5 seconds.

Images