JP2020059302A - Course control method for automated driving - Google Patents

Abstract

To provide a course control method for automated driving, which allows for driving in a condition where a curbstone on a road is detected by a measurement device such as LiDAR even when a GPS is not available and which can control with high accuracy a clearance between a curbstone and a bus when the bus is driven alongside a bus stop.SOLUTION: A course control method for automated driving includes a target route formulas that can conform to the landform at an actual place, detects a lateral deviation and an angle deviation to a target route from the shape of a curbstone such as a bus stop with a measurement device such as LiDAR, and calculates a steering angle for the target route depending on a vehicle model and thus allows a vehicle to travel along the target route and stop at a predetermined location.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a route control method in automatic driving using a distance measuring device LiDAR (Light Detection and Ranging) using laser light.

Patent Document 1 proposes a method of making a gap between a curb and a parallel to a curb as small as possible when a bus (vehicle) is stopped at a bus stop by automatic driving.
In this patent document 1, the locus drawn by the rear axle side wheels is used as the root locus representing the movement of each part of the vehicle when approaching the bus stop and stopping the bus while maintaining the lane.

  Patent Document 2 proposes an automatic driving method (platooning) using an optical distance measuring device. In Patent Document 2, the position of the front vehicle is detected by an optical distance measuring device, the direction and distance to the specific portion of the front vehicle are detected, and the front wheel actual steering angle that follows the locus to the specific portion is calculated. ing.

  In Patent Document 3, when LiDAR is used to perform scanning with laser light in the horizontal direction and the vertical direction to measure the position and shape of an object existing around the vehicle, the influence of a complicated shape of a road boundary and Due to the influence of noise, targeting may be performed so that the target frame may straddle the lane boundary line even though it is a roadside object outside the lane boundary line. In order to prevent this, it has been proposed to provide a lane boundary line recognizing means, an in-lane area extracting means, and a judging means.

  Currently, research on automatic driving utilizing a speed / position recognition function by GPS is progressing. Demonstration experiments of automatic driving using VRS-GPS, which can obtain accuracy of several cm by receiving position information from GPS satellites and correction information using wireless communication, are in progress. VRS stands for Virtual Reference Station and is called a virtual reference point in Japanese.

GPS has an error called multipath. Multipath occurs on various obstacles (buildings, trees, power lines, ground, etc.). When the multipath occurs, an error occurs in the positioning accuracy, the automatic driving route is disturbed, and the route maintaining traveling cannot be performed.

  On the other hand, the above-mentioned LiDAR has improved performance, and a commercialization plan for autonomous driving that can measure a distance from a distance of 120 m with an accuracy of less than 2 cm is reported. In a place where GPS cannot be used by utilizing the speed / position recognition by GPS, a technology for automatically traveling by measuring the distance to the curb etc. by LiDAR can be considered.

Japanese Patent No. 5981010 Japanese Patent No. 6109998 JP, 2008-92483, A

  It is an object of the present invention to enable a rider to perform a task assigned to an autonomous driving vehicle such as running or landing at a bus stop by detecting a road curb or a bus stop curb even when GPS cannot be used. In a general road including a tunnel, in order to detect a curb and travel along the curb, it is necessary to detect the parallel and the distance between the vehicle and the curb and perform steering control along the curb. In order to arrive at the bus stop, the bus stop (Bus bay) is required to dock with a minimum clearance without climbing a curb according to the shape of the bus (Docking).

  However, Patent Document 1 describes the geometry of the target locus and the steering control for the normal stop at the bus stop, and does not refer to the method and means for confirming the clearance between the curb and the vehicle body.

  Patent Document 2 is concerned with vehicle position recognition regarding following of a preceding vehicle, and does not refer to automatic driving by detecting a curb.

  Patent Document 3 relates to narrowing down the target around the vehicle detected by LiDAR to within the vehicle lane area only, and does not relate to running or normal wearing after detecting a curb.

Targets that set the distances between the curbstones, guardrails, white lines, fences, and other targets along the route based on maps (including digital maps) and the vehicle body in advance, and that enable running while maintaining the set intervals The route is created by a curve formula or a number table, the own vehicle position is confirmed by LiDAR during traveling, the deviation between the own vehicle position and the target route is calculated, and steering is performed to correct this deviation.
The deviation is detected by monitoring the distance and angle between the curb and the vehicle body by LiDAR.

  It has a target route formula that can match the actual terrain, has a vehicle model that determines the rudder angle along the target route, and determines the vehicle attitude along the curb by determining the curb gap along the target route and checking with LiDAR Then, since the correction steering is performed, it is possible to accurately travel the target route and stop at the correct position of the bus stop.

In particular, in autonomous driving using GPS, it operates in forests, roads, cliffs, under the net, in high-rise buildings, where GPS positioning accuracy decreases, and in tunnels, underground spaces, indoors where GPS cannot be used. It is possible to improve the accuracy of porting to a bus stop or cargo handling platform.

Relationship diagram between rear axle trajectory and vehicle front edge trajectory when laying on curb Explanatory drawing of how to take the target trajectory to be laid on the curb Explanatory drawing of detection of curb gap and vehicle posture by LiDAR Explanatory drawing of detection of curb gap and vehicle posture by LiDAR Illustration of the actual steering angle calculated from the rear-axis trajectory Illustration of control flow chart

Embodiments of the present invention will be described below with reference to FIGS.
FIG. 1 is a diagram showing a relationship between a rear axle locus and a vehicle front end locus when horizontally laid on a curb. Shows the rear axle path from (A) to exit (B).

The extremely low speed range, such as when the lateral association the curb, because there is no slippage of the tire, R1 point and the vehicle front end F1 point corresponding to the rear axle pivots about a O ₁ point on the rear axle extension together. Similarly R2-point and the vehicle front end F2 points corresponding thereto is pivoted about the O ₂ points on the rear axle extension together. Therefore, if there is a curb on the turning trajectory of the F2 point, the F2 point will ride on the curb. Eventually, the car body will be stopped parallel to the curb at the R3 and F3 points, but it will be necessary to devise a way to take the trajectory to reach R3 and F3 without eliminating the curb riding at the F2 point.

  FIG. 2 shows how to take a target locus to be laid horizontally on the curb without riding on the curb. (1) First, make a deep cut to bring the front roadside edge of the vehicle closer to the curb. (2) While gradually returning the handle, move forward until the curb gap is parallel to the front and back. The trailing axis locus at this time can be approximated by the exponential function of the equations (1) and (2).

  The front roadside edge of the vehicle moves to F1, F2, F3 and the curb. A means for preparing a target locus for a route from the position of the vehicle to the stop position before starting the movement and for confirming that the target locus is followed is required. LiDAR is used in the present invention as a means for confirming that the target trajectory is being followed.

3 and 4 are explanatory diagrams for detecting the curb gap and the vehicle posture by LiDAR. Here, FIG. 3 is a case where the curb is a straight line, and FIG. 4 is a case where the curb is a curve (including a broken line). By equipping the front center of the vehicle with LiDAR, detecting A, B, and C, and grasping the distance D to the curb and the attitude angle Φn of the vehicle with respect to the curb, it is possible to confirm that the target trajectory is being followed. . Here, A is an intersection of a line extending from the vehicle front left end point ( _PFE ) to the vehicle front and the curb, and B is a line extending from the vehicle front left end point ( _PFE ) to the vehicle width direction and the curb. The intersection, C, is the distance between A and B.

FIG. 5 shows an explanatory diagram of the actual steering angle calculated from the rear-axis trajectory. We will explain how to steer when setting a target locus in the terrain of the site and following the locus.
The target locus is given by the equations (1) and (2), and the curvature of this equation is given by the equation (3).

Here, f (x) is equation (1).
The front wheel actual steering angle for drawing this curvature is given by equation (4).

Here, δ is an actual front wheel steering angle, and K _SF is a stability factor, which is a characteristic value expressing the change characteristic of the turning radius with respect to the square of the vehicle speed when the vehicle travels on a curve. v is the vehicle speed, ρ is the curvature, and L (lowercase in the equation) is the wheelbase.

In the upper part of FIG. 5, the solid line is the locus of the rear shaft, and the dotted line is the locus of the corresponding central portion of the front end of the vehicle. The lower row is the actual steering angle. The processes (1), (2) and (3) are observed in the change of the actual steering angle.
(1) First, make a large cut to bring the front end of the vehicle body close to the curb. (2) Cut back and move forward while laterally moving the front end of the vehicle body. (3) Gradually cut back like tracing a curb at the front end of the car body and stop with the curb and the car body parallel. This process is consistent with a person's practice and improved steering.

If the steps (1), (2) and (3) are to be accommodated in a short distance, it is done by increasing λ in the equation (1). In that case, the locus of the front end of the vehicle rises as shown by (4) in the figure, and the curb rides up in FIG.
If the curb is cut according to this riding shape and the curb is constructed, it will also be a bus stop adapted to the site space.

  FIG. 6 shows a control flowchart. The steps of a task of creating a target locus for laying it down at a bus stop and joining the target locus from the vehicle position to the target locus are shown.

  In step (1), the latitude / longitude / azimuth of the orbit of the route key point is acquired and interpolated by a method such as a clothoid curve or a least squares method to prepare the target trajectory in advance. If it is a bus stop, prepare a curve formula, a number table, etc. for the running locus to be placed alongside the bus stop. Information such as latitude / longitude, curvature, gradient, and control speed is given by attaching an ID.

In step 2, the target locus and the site are aligned. Align the relative axes of the earth coordinate system and the vehicle coordinate system at the starting point and transit point (match the vehicle's position and orientation with landmarks (bus stops, stop lines, curbs, etc.).

  In step 3, on the spot such as a bus stop, LiDAR is used to grasp A, B, C, D and Φn in FIG.

  In step 4, the position coordinates and direction of the vehicle are recognized. Check the location of your vehicle in light of the orbital and numerical tables at key points such as bus stops.

  In step 5, the target point coordinates and orientation on the target trajectory are recognized. Determine how many meters ahead the target point will be set at the trial operation stage. Normally, the distance is 4 m (common sense for the forward-looking model).

  In step 6, the control rudder angle is determined. The actual steering angle is determined according to the steering angle calculation formula based on the curvature of the interpolation curve and the vehicle speed by performing clothoid interpolation or the like from the coordinates and direction of the own vehicle and the coordinates and direction of the target point.

At step 7, the target point is reached. If there is a bus stop, stop there. If it is a general road, the deviation of the own vehicle position error is grasped, the process is returned to the step (4) and the task is ended while repeating the steps (4) to (7).

Claims (3)

  Set the distance between the vehicle body and the curbstone, guardrail, white line, fence, etc., which are provided along the operation route based on the map, and the target route that allows traveling while maintaining the set interval is a curve type or number table It is characterized in that the vehicle position is confirmed by a measuring device using laser light during traveling, the deviation between the own vehicle position and the target route is calculated, and steering is performed to correct this deviation. Course control method in automatic driving. In the route control method for automatic driving according to claim 1, the locus is calculated by the formula (1) and the formula (2) for laying on a curb such as a bus stop, and the locus is followed by the formula (3) and the formula (4). A route control method in automatic driving, characterized by calculating an actual steering angle for driving.

  The route control method in automatic driving according to claim 1 or 2, wherein the measuring device detects a lateral deviation and an angular deviation from a curb shape of a bus stop or the like to calculate an actual steering angle for following the trajectory. And a route control method in automatic driving.

Images