JPS62274316A - Obstacle detecting device for unmanned vehicle - Google Patents

Abstract

PURPOSE:To facilitate a process for determining the operation of an unmanned vehicle to an obstacle and to reduce the scale of a processor by detecting and computing the position coordinates of the obstacle. CONSTITUTION:Respective sensors of a distance measuring means 1 send out a corresponding signal to the obstacle when the obstacle is within a measurable range. This means 1 covers a necessary range of an obstacle detection range by the whole measurable area of the sensors and divides the whole area into plural small overlapping detection areas of only one or plural sensors. Then, when a sensor detects the obstacle, an area deciding means 2 decides the sensor and decides the sensor installation position in a storage means 4 and a small area where the obstacle is present on the basis of the shape value of the measurable area. Further, an arithmetic means 3 computes the coordinates of the obstacle position in a coordinate system fixed to the unmanned vehicle in the small decided area on the basis of a distance value from the sensor, the installation position of the sensor, and the shape value of the measurable area.

Description

[Detailed Description of the Invention] 3. Detailed Description of the Invention [Field of Industrial Application] The present invention relates to an obstacle detection device for an unmanned vehicle, and particularly to a device for detecting obstacles in an unmanned autonomous vehicle. The present invention relates to an obstacle detection device for detecting obstacles.

[Conventional technology]

Conventionally, as an obstacle detection device for an unmanned vehicle, several non-contact obstacle detection sensors are installed side by side in the left and right direction at the front of the unmanned vehicle, and the distance value obtained from these sensors to the obstacle is used to A method has been proposed for detecting the approximate location of an obstacle based on the above-described method (Japanese Patent Application Laid-open No. 74905/1983).

As shown in FIG.
Ultrasonic sensor 5 as a non-contact obstacle sensor on the front of the
A and 5B are arranged side by side in the width direction of the vehicle. The ultrasonic sensors 58 and 5B each have the same configuration, and transmit ultrasonic waves intermittently over a predetermined range in front of the vehicle body D1, and receive reflected waves from the obstacle A during the intermittent transmission. By doing so, it senses the obstacle and detects the distance to the obstacle $JA based on the time required for transmitting and receiving ultrasonic waves. Then, the ultrasonic sensors 5A and 5B are arranged so that the obstacle sensing area can be divided into three parts and detected.
Ultrasonic sensors 5A and 5B are arranged so that their respective obstacle sensing areas Xs and Xt partially overlap.

Therefore, by combining the obstacle detection results of the two sensors 5A and 5B, it is possible to determine in which of the three sensing areas A+ and At-As the obstacle A is present. That is, as shown in the sensing position discrimination table of FIG. 2, if only the sensor 5A senses an obstacle, it will be placed in the front left side of the vehicle body, that is, in the sensing area A1, and if only the sensor 5B senses an obstacle, it will be placed in the front right side of the vehicle body, that is, in the sensing area A1. In sensing area A3,
When both sensors 5A and 5B detect an obstacle, it is determined that the obstacle exists in front of the center of the vehicle body, that is, in the sensing area A:.

In addition, as shown in FIG. 3, when one or more narrow-directivity ultrasonic range finders 100 are rotated and an obstacle A is detected, the angle of the range finder at that time and the distance to the obstacle are determined. A device has been proposed that calculates the position of an obstacle as a vector quantity B from the distance (Japanese Patent Application Laid-Open No. 58-203518). This device can detect the distance to obstacles around an unmanned vehicle. A device characterized by having a detection means in which one or more detectors are arranged at equal or unequal intervals on the circumference, and a function to rotate this detection means at an angle within a certain range. Then, the distance and direction from the unmanned vehicle to surrounding obstacles are expressed as vector quantities and output.

[Problem that the invention seeks to solve]

However, the obstacle detection device disclosed in Japanese Unexamined Patent Publication No. 59-74905 detects the position of the obstacle by representing the distance from the unmanned vehicle and the approximate direction of left, right, and center. Since the location of an object is not expressed in coordinates,
It is difficult to process data such as associating it with past detection positions. As a result, sufficient information about obstacles cannot be obtained, and the only way to do so is to perform detour operations using fixed steering operations or avoidance operations based on trial and error, or to be unable to respond when the distribution of obstacles is somewhat complex. There was a point.

Furthermore, in the method of detecting obstacles by rotating an ultrasonic distance meter such as the obstacle detection device disclosed in Japanese Patent Application Laid-open No. 58-203518, it takes time to detect an obstacle because rotational scanning is performed. In addition to taking too much time, the amount of information increases because it also incorporates distance values of surrounding obstacles that are unrelated to the direction in which the unmanned vehicle is traveling, making the processing for selecting the information necessary for the unmanned vehicle to run complicated. There was a point.

When the present inventors investigated the cause of the main problem with the conventional device, it was found that it was caused by not detecting the position of an obstacle as a coordinate value of a single point. The inventors of the present invention have therefore reached the conclusion that the main problems of the prior art can be solved by using means for detecting and calculating the position coordinates of obstacles. As a method for detecting and calculating the position coordinates of an obstacle, there is an area determining means that limits the range in which the obstacle exists to a small area and makes it easy to derive the position of the obstacle, and a method that makes it easy to derive the position of the obstacle within this area. We decided to use an obstacle coordinate calculation means that estimates and calculates the obstacle position coordinates in the coordinate system fixed to the unmanned vehicle.

SUMMARY OF THE INVENTION An object of the present invention is to provide an obstacle detection device that solves the problems of the prior art, has a simple device configuration, and can calculate the position coordinates of an obstacle by detecting its distance. do.

[Means for solving problems]

In order to achieve the above object, the present invention provides an obstacle detection device for an unmanned vehicle that autonomously moves along a preset travel course, as shown in FIG. At least two sensors are provided to measure the distance to an obstacle existing within a measurable area smaller than the range where The entire measurable area can be divided into a small area smaller than the measurable area and measurable by one sensor and a small area smaller than the measurable area and measurable by at least two sensors. Distance N measuring means 1 in which the sensor is installed; storage means 4 for storing in advance the installation position of the sensor and the shape value of the measurable area; The installation position of the sensor that determines whether the obstacle has been detected and the measurement times P stored in the storage means 4: an area determination means 2 that determines a small area in which an obstacle exists based on the shape value of the area; Based on the installation position of the sensor and the shape value of the measurable area stored in the means 4, the position of the obstacle within the small area determined by the area determining means 2 is estimated and the position of the obstacle is detected in the coordinate system fixed to the unmanned vehicle. Obstacle coordinate calculation means 3 for calculating object position coordinates
It is characterized by having the following.

[For production]

Next, the present invention will be explained in detail. In FIG. 4, each sensor of the distance measuring means 1 outputs a signal corresponding to the distance to the obstacle when the obstacle exists within the measurable area. The distance measuring means 1 covers the necessary range for detecting obstacles when an unmanned vehicle is running by using the entire measurable area of the sensor, and also measures the entire measurable area using only one sensor. The measurable areas of the respective sensors are arranged so as to have a certain shape with respect to each other so that the measurable areas of the respective sensors are divided into small areas such as possible small areas or small areas where the measurable areas of several sensors overlap. When a sensor detects an obstacle, the area determination means 2 determines which sensor has detected the obstacle, and also stores the sensor installation position and measurement times P: a shape value of the area in a storage means 4 in advance. Based on the stored installation position and the shape value of the measurable area, it is determined which small area is the area in which the obstacle exists. Moreover, the obstacle coordinate calculation means 3
Within the small area determined by the area determining means 2, the area fixed to the unmanned vehicle is based on the distance value output from the sensor and the sensor installation position and shape value of the measurable area stored in the storage means 4 in advance. Calculate the obstacle position coordinates in the coordinate system.

(Effects of the Invention) As explained above, in the present invention, the position of the obstacle is represented by coordinates, so the information on the obstacle is expressed in an accurate and concise manner. When calculating these coordinates, the measurable area is divided into small areas and the small area where the obstacle exists is determined, thereby limiting the range where the obstacle exists to the small area. Therefore, it is possible to calculate the coordinates of obstacles with high accuracy.

Furthermore, since a plurality of sensors can be processed in parallel at the same time, it is possible to perform obstacle detection processing in a shorter time than in the case where the sensors are rotated to detect obstacles.

[Explanation of B camphor]

The present invention may take the following embodiments when implemented.

In the first Li, when the sensor detects an obstacle, the area determining means determines which sensor has detected the obstacle, and also determines the installation position and measurement of the sensor stored in the storage means. The present invention is characterized in that a small area in which an obstacle exists is determined based on a shape value of the possible area and a distance value output from a sensor that has detected the obstacle. In this first aspect, in the judgment operation of the area judgment means, in addition to which sensor has detected the obstacle, the distance value output from that sensor is used to predict the presence of an obstacle that satisfies the condition. A method is used to determine the small area that will be used. With this aspect, it is possible to further divide the area into smaller areas, including division in the distance direction, and determine the area where the obstacle is expected to exist, thereby improving the accuracy of judgment by limiting the area where the obstacle is present even smaller. can be done.

Further, in a second aspect, the obstacle coordinate calculation means detects an obstacle within the small area determined by the area determination means based on the installation position of the sensor and the shape value of the measurable area stored in the storage means. The object is fixed to the unmanned vehicle by estimating the position of the object, assuming a representative line within the small area, and calculating the coordinates of the point corresponding to the distance value output from the sensor that detected the obstacle on the representative line. It is characterized by calculating the position coordinates of an obstacle in a coordinate system. In this second case a, as the obstacle position calculation method of the obstacle coordinate calculation means, a representative line whose coordinate values are known is assumed within the small area where the determined obstacle exists. . and,
A method is used in which the position coordinates of the obstacle are calculated by approximating that the obstacle exists at the position of the distance value output by the sensor and calculating the coordinates on the representative line corresponding to the distance value.

The representative line assumed for the small area can be selected based on criteria such as minimizing errors due to approximation. With this aspect, the position coordinates of the obstacle can be obtained with relatively high accuracy through simple calculations.

And the third L! For Mr. i, when the area determination means determines that an obstacle exists in a part where the measurable areas partially overlap, the obstacle coordinate calculation means calculates the distance value output from the sensor that detected the obstacle. and the installation position of each sensor stored in the storage means, the obstacle position is calculated by calculating the coordinates of the intersection of a circle whose center is the installation position of each sensor and whose radius is each distance value. It is characterized by calculating coordinates. In this third aspect, when the area determination means determines that the obstacle exists in a portion where the measurable areas partially overlap, the obstacle coordinate calculation means is configured to calculate When it is determined that an obstacle exists, the position coordinates of the obstacle are calculated based on the distance values from at least two sensors and the installation position of the sensor stored in advance in the storage means. According to this aspect, the coordinates of the point can be calculated without using approximation, so the accuracy of determining the position of the obstacle is improved.

〔Example〕

Embodiments of the present invention will be described in detail below with reference to the drawings.
The obstacle detection device for an unmanned vehicle of this embodiment is an application of the present invention to a collision avoidance device for an autonomous unmanned vehicle that does not use a guide line.

FIG. 5(1) is a block diagram showing the configuration of a collision avoidance device according to an embodiment of the present invention.

The device of this embodiment belongs to the above-mentioned aspect 1.2.3, and is a driving course storage means that stores information on the traveling course such as a target course and target speed in advance! 1, a position/azimuth measuring means (1) for measuring the current position/azimuth of the vehicle, the position/azimuth of the vehicle obtained from the position/azimuth measuring means (1), and the target course and target speed stored in the driving course storage means 11. A driving control means (■1) calculates the amount of restraint required to drive the unmanned vehicle based on the amount of restraint required to run the unmanned vehicle, and a driving means (actually driving the unmanned vehicle) based on the restraint output from the traveling control means (■1). ■1, obstacle detection means ■1 that detects the position of the obstacle, and obstacle detection means ■1 that processes the obstacle information obtained from the obstacle detection means ■1 to determine the necessity of avoiding the obstacle. Obstacle information processing means to detect the distribution state of objects■
1, and obstacle storage means for storing position information of obstacles ■1
When an obstacle to be avoided is detected, the driving control means is instructed to decelerate, and a detour to avoid collision with the obstacle is calculated, and the detour is set as the course for future travel. It is composed of a storage means (1) and an avoidance operation determining means (1) stored in the storage means (1).

Obstacle detection means ■1 includes two ultrasonic sensors 51R and 51L that are capable of transmitting and receiving, which are attached to the left and right ends of the front part of the car facing forward in the driving direction, and the respective sensors 51R and 51L.
The presence or absence of an obstacle is determined from the presence or absence of ultrasonic waves received by the ultrasonic wave transmitting circuit 52 and the sensors 51R and 51L, and when an obstacle exists, the delay from the time of transmission of the reflected wave is calculated. Distance measurement circuit 53R153L that calculates the distance to the obstacle that reflected the ultrasonic wave on the left and right sides.
A distance measuring means 1 is provided. In addition, the obstacle detection means (■) includes a left and right distance measuring circuit 5.
3R, 53L captures (detects) the obstacle at a distance #dl, d, and determines in which small region the obstacle exists; Within the small area determined by means 2, the distance measuring circuit 53R
Obstacle coordinate calculation means 3 that calculates the obstacle position coordinates in the coordinate system fixed to the unmanned vehicle based on the distance value from 153L, and sensors necessary for the processing of the area determination means 2 and the obstacle coordinate calculation means 3. A storage means 4 is provided for storing the installation position and the shape value of the measurable area.

Figure 6 shows the installation positions of the ultrasonic sensors SIR and 51L attached to the front of the unmanned vehicle 90, the shapes 91R and 91L of the measurable areas of the individual sensors, and the measurable areas determined by the shapes 91R and 91L. FIG. 6 is a diagram showing small regions (1) to (2) that can limit the range in which obstacles exist. By appropriately setting the installation position of the ultrasonic sensor and the shape of the measurable area of each sensor, the entire necessary range necessary to measure an obstacle can be obtained by measuring the area where the measurement times P and 95 areas of the two sensors overlap. ■, a small area detected by only one sensor and divided in the front and back direction according to the distance value, ■, and a small area ■, ■
It is divided into five small areas.

The area determining means 2 determines in which small area of the aforementioned Φ to ■ the detected obstacle exists.
), comparators 01 to C10 and AND circuit A1
~A8 and arithmetic circuits E1 and E2.

The comparator C1 receives the distance value d from the left distance measuring circuit 53L.
1 and a pre-given threshold value 111 (FIG. 6), di<1. The logical signal 0. is l if djl≧11 and is O if djl≧11. and outputs a logic signal 1nvo+ which is opposite to the logic signal 01.

In other words, in the comparator CI, the left distance R'/M constant circuit 53L
If the obstacle detected in is present in the area (3), the logic signal O8 becomes 1. Comparator C2 is the distance measuring circuit 53 on the right side.
The distance value d1 from R is compared with a predetermined threshold value 11, and d,<1. In the case of l5dj! ≧11. Logic signal 0 which becomes O in case of ! and outputs a logic signal inv Oz which is opposite to the logic signal 0□. That is, comparator C
2, the logic signal 02 becomes 1 when the obstacle detected by the right distance measuring circuit 53R exists in the area (3).

AND circuit A1 outputs logic signal 1nvo+ and logic signal 0!
are input, the AND is performed, and a logic signal S is output.

When the logic signal S is 1, the area 3 is selected as the area where the obstacle exists. This means that there are no obstacles in the area
(1nvo+=1), and there is an obstacle in area (Of-1).

AND circuit A2 outputs logic signal 01 and logic signal inv O
z and performs a logical AND operation to obtain a logical signal S! Output. When the logic signal St is 1, area (2) is selected as the area where the obstacle exists. This means that there are no obstacles in the area
(1nvoz = 1), and there is an obstacle in the area (Os"1).The AND circuit A3 receives the logic signal o1 and the logic signal o2, performs the AND operation, and outputs the logic signal S.

Output. In the AND circuit A3, when an obstacle is present in both the area Φ and the area (2), the logic signal S becomes l when <at -Ox -1). When the logic signal S, is I, the comparator C3 is activated.

The comparator C3 receives the distance value d from the left distance measuring circuit 53L.
It is compared with the distance value d1 from the right distance measuring means 53R, and a logic signal 0. and logic signal 0. It outputs a logic signal 1nvos that is opposite to . Logic signal 0. but! In this case, area (2) is selected as the area where the obstacle exists, and when the logic signal inv 02 is I, area (2) is selected as the area where the obstacle exists. In other words, this comparator c3 is activated when an obstacle exists in both area ■ and area ■, compares the distance from each sensor to the obstacle, and selects the area with the smaller distance as the area where the obstacle exists. Select as.

AND circuit A4 has logic signal 1nvC)+ and logic signal 1
nvot is input, a logical AND is performed, and a logic signal S4 is output. When logic signal s4 is 1, comparator C4 and comparator C5 are activated. In AND circuit A4, if the obstacle does not exist in either area ■ or area ■ (1nvo+
-1nvoz -1), the logic signal S4 becomes 1. When logic signal s4 is 1, comparator C4 and comparator C5 are activated.

The comparator C4 receives the distance value d from the left distance measuring circuit 53L.
1 and a pre-given threshold value Is (Fig. 6), di<I! ,In Case of! , (If Ml≧l, outputs logic No. 13 o4 which becomes 0, and logic signal inv Oa which is opposite to logic signal 04.

That is, the comparator C4 determines whether or not the distance value obtained by the left distance measuring circuit 53L is less than or equal to the sensor's maximum detection path #13, and whether or not the sensor has detected an obstacle. If so, the logic signal 04 is set to 1. The comparator C5 receives the distance value d from the right distance measuring circuit 53R and a predetermined threshold value 1. Compare d,
<1. In the case of l, dj! ≧l, the logic signal O1 and the logic signal O5, which become O, are the opposite logic signal inv.
Outputs O5. That is, the comparator C5 determines whether the sensor has detected an obstacle when the distance value obtained by the right distance measuring circuit 53R is less than or equal to the maximum detection part M1s of the sensor, and determines whether the sensor has detected an obstacle. If so, the logic signal 0. of! Make it.

The AND circuit A5 inputs the logic signal 04 and the logic signal 0°, performs an AND operation, and outputs a logic signal S. When the logic signal S is l (on = os -1), both sensors are detecting an obstacle, but there is no obstacle in the area ■ and the area ■. When the logic signal S, is 1, the comparator C6 is activated.

The AND circuit A6 inputs the logic signal 1nvO4 and the logic signal O3, performs an AND operation, and outputs a logic signal S. When the logic signal S is 1 (1 nvo, 0s = 1), only the right sensor is detecting an obstacle, and there is an obstacle outside the region (2). When logic signal Sh is 1, comparator C7 is activated. The AND circuit A7 inputs the logic signal o4 and the logic signal inv Os, performs an AND operation, and outputs a logic signal S. When the logic signal S is 1 (04m 1nvos = 1), only the sensor on the left side is detecting an obstacle, and there is an obstacle outside the area ■. When the logic signal S, is 1, the comparator C8 is activated. The AND circuit A8 inputs the logic signal tnvo4 and the logic signal invO5, performs an AND operation, and outputs a logic signal S. When the logic signal S is 1 (1nvo
a −1nvos = 1) is a case where neither sensor on both sides detects an obstacle, and there is no obstacle.

Logic signal S! When , it is determined that the obstacle does not exist.

The comparator C6 receives the distance value d from the left distance measuring circuit 53L.
1 and the distance value d1 from the right distance measuring circuit 53R, and dj! Logic signal 0. is 1 when <d, and 0 when d1≧d. The comparator C6 outputs a logic signal 1nvoa which is opposite to the logic signal Oh. Both sensors detect an obstacle, but when there is no obstacle in the area ■ and area ■, the left and right sensors Comparing the distance values, if the distance 41idl on the left side is small, the logic signal O4 becomes l, and if the distance value d7 on the right side is small, the logic signal inv
O4 becomes 1. When the logic signal Oh is 1, the comparator C
When IO is activated and logic signal 1nvoa is 1, comparator C9 is activated.

The comparator C7 receives the distance value d from the right distance measuring circuit 53R.
, and a pre-given threshold value 12 (FIG. 6), d,<1. In the case of l, d.

Logic signal 0. which becomes O when ≧i. and logic signal 0. It outputs a logic signal inv Oq which is opposite to the logic signal inv Oq .

Logic signal 0. When is l, area ■ is selected as the area where the obstacle exists, and when the logic signal inv Oq is 1,
It is determined that there are no obstacles. Comparator C7 detects an obstacle when only the right sensor is detecting an obstacle and there is an obstacle in area other than area ■, or when both sensors are detecting an obstacle but the obstacle is in area ■. If the obstacle does not exist and the distance value d1 of the right sensor is smaller, it is determined that the obstacle is in the area ■ by comparing the threshold value 18 indicating the maximum distance of the area ■ with the distance value d on the obstacle. ■Does it exist?
Threshold value 12 indicating the maximum distance of the obstacle and distance value d1 on the obstacle
This is determined by comparing the
If it is smaller than that, it is determined that an obstacle exists in area (3).

Comparator C8 calculates the distance M41 from the left distance measuring circuit 53L.
idl and pre-given threshold? +i J & and dj<J! ! In the case of 1Sdl≧12, the logic signal O1 becomes 0 and the logic signal 0. It outputs a logic signal 1nvos which is opposite to , and when the logic signal 0° is l,
When region (2) is selected as the region where the obstacle exists, and the logic signal invow is 1, it is determined that no obstacle exists. Comparator C8 detects an obstacle when only the sensor on the left side detects an obstacle and there is an obstacle outside area ■, or
Both sensors are detecting an obstacle, but the obstacle is in the area.
If the distance value d1 of the right sensor is smaller, the distance value d1 above the obstacle is compared with the threshold value 12 indicating the maximum distance of the area ■, and the obstacle is located in the area ■.
If di is smaller than 12, it is determined that an obstacle exists in area (3).

The comparator C9 receives the distance value d from the left distance measuring circuit 53L.
1 and the output value f, (d,) when the distance value d from the right distance measuring means 53R is input to the arithmetic circuit E1, and if d1≦[1(d,), 1. , al>r, (
dr), the logic signal 0. and logic signal 0
．． Outputs logic signal 1nvoq opposite to logic signal O9
When is 1, region 3 is selected as the region where the obstacle exists, and when logic signal inv O* is 1, comparator C7 is activated. Comparator C9 detects an obstacle when both sensors detect an obstacle, but there is no obstacle in area ■ and area ■, and the distance values of the left and right sensors are compared, and if the distance value d on the right side is small, it is determined that there is an obstacle. The distance value d on the right side is used as a threshold value for determining whether or not is in the area ■, and is used in the calculation circuit E.
If Sat, which is the value obtained by using the output value fl(d,) when inputting 1, is less than or equal to [1(d,), area (2) is selected as the area where the obstacle exists.

The comparator C10 calculates the distance # from the right distance measuring circuit 53R.
(di d , and the output value f when the distance value d1 from the left distance measuring circuit 53L is input to the arithmetic circuit E2, (d
Logic signal 01° which becomes 0 when a, ≦f+ (dl), and 0 when d, ≦f, (dl) and the logic opposite to the logic signal 01°. Signal 1nvo+s
When the logic signal 01° is 1, the area ■ is selected as the area where the obstacle exists, and the logic signal 1nvo+
. When is l, comparator C8 is activated. This means that both sensors are detecting obstacles, but there are no obstacles in area Φ and area ■, and if the distance values of the left and right sensors are compared and the distance value d1 on the left side is small, the obstacle is in area ■. The output value fg when the left distance value d1 is input to the arithmetic circuit E1 as a threshold value for determining whether
(djl). d, is rt (dj
In the following cases, the area ■
is selected.

Arithmetic circuits El and E2 calculate the human power value by d (=d, , d
l), output value r+ (f+ (d,), r+
(dl)), the calculation shown in the following equation <2-1> is performed. This equation is used to calculate a threshold value for determining whether there is an obstacle in area (2). Here, W indicates the distance from the center point of the front of the vehicle to the two ultrasonic sensors 51R151L attached facing forward on the left and right sides of the front of the vehicle, as shown in FIG. As shown in FIG. 6, the inclination angle of the line segment indicating the edge of the obstacle detection area of each sensor with respect to the traveling direction of the vehicle is shown.

f l =2w+sin φ + (d"-4
11+ cos” φ) ・−・ −・<2-1>
As shown in FIG. 5(3), the obstacle coordinate calculating means 3 is comprised of five coordinate calculating circuits P1 to P5 activated according to the determination result of the area determining means 2. Each coordinate calculation circuit includes an X coordinate calculation circuit pH, P21. P31°Pd2. P51 and Y coordinate calculation circuit P12. P22゜P32. P42. Consists of P52. The five coordinate calculation circuits P1 to P5 correspond to the five regions (1) to (2) determined by the region determining means 2.

The X coordinate calculation circuit pH of the coordinate calculation circuit Pi receives the distance value d1 from the right distance measurement circuit 53R, and performs the calculation shown by the following equation.

C) +-drcos ρ ・・・ ・・・ ・・・
... ... ... ... <2-2> The Y coordinate calculation circuit P12 of the coordinate calculation circuit P1 inputs the distance value d from the distance measurement circuit 53R on the right side, and performs the calculation shown by the following formula. conduct.

Oy = W+ -d, sin ρ
・・・・・・ ・・・ ・・・ <2-3> Above <2-
2> and <2-3> In the region ■ where formulas are calculated, the second aspect of the obstacle position calculation method is used, and the line segments (representative lines) assumed within the region are as shown in Figure 6. , right sensor 51
This is a center line R1 drawn from R to the region. ρ indicates one angle of this line segment with respect to the direction of travel of the car (ρ〉0)
, and W indicate the distance from the center point of the front of the vehicle to the two ultrasonic sensors 51R and 51L, which are attached to the front left and right of the vehicle facing forward.

The final coordinate calculation circuit P21 of the coordinate calculation circuit P2 inputs the distance value dl from the left distance measurement circuit 53L, and performs the calculation shown by the following equation.

Ox snow d1 ・CO8ρ ・・・・・・・・・
... ... ... <2-4> Coordinate calculation circuit P
The Y coordinate calculation circuit P22 of No. 2 is the distance measuring circuit 53L on the left side.
Input the distance value di from , and perform the calculation shown in the following equation.

Q, = −W, + d1・s
in o--<2-5> above <2-4>, <2-5>
In the area ■ where the formula is calculated, the second 8-way obstacle position calculation method is used, and the line segment assumed within the area is the center line drawn from the left sensor to the area as shown in Figure 6. It is L1. ρ indicates the inclination angle of this line segment with respect to the direction of travel of the vehicle (ρ>Q).

The X coordinate calculation circuit P31 of the coordinate calculation circuit P3 inputs the distance value d1 from the right distance measurement circuit 53R, and performs the calculation shown by the following equation.

The Y-coordinate calculation circuit P32 of the coordinate calculation circuit P3 receives the output value OX of the X-coordinate calculation circuit P31 and performs calculations expressed by the following equation.

07 Snow Ox -tanV ・・・ ・・・
・・・ ・・・ ・・・ ＜2-7＞Above <2
-6> and <2-7> In area ■ where formulas are calculated, the second aspect of the obstacle position calculation method is used, and the line segment assumed within the area is from the center O between the seven sensors to area ■. Pulled center iR2
It is. '1' indicates the inclination angle of this line segment with respect to the direction of travel of the vehicle (MP>O).

The Y coordinate calculation port 1P42 of the coordinate calculation circuit P4 inputs the distance value d from the distance measurement circuit 53R on the right side and the distance value dl from the distance measurement circuit 53L on the left side, and performs the calculation shown by the following equation.

The X coordinate calculation circuit P41 of the coordinate calculation circuit P4 calculates the distance value d7 from the right distance measuring circuit 53R and the Y coordinate calculation circuit P4.
The output value Ov of 42 is input, and the calculation shown by the following equation is performed.

The third aspect of the obstacle position calculation method is used in the region (■) where the above formulas <2-B> and <2-9> are calculated, and each distance value is calculated with the set position I of each sensor as the center. The position coordinates of the obstacle are calculated by finding the intersection on the X axis (FIG. 6) of a circle whose radius is .

The X coordinate calculation circuit P51 of the coordinate calculation circuit P5 manually inputs the distance value d1 from the left distance measurement circuit 53L, and performs the calculation shown by the following equation.

The Y-coordinate calculation circuit P52 of the seed calculation circuit P5 receives the output value OR of the X-coordinate calculation circuit P51 and performs the calculation shown in the following equation.

Ov ” OK -tan 'P...
......<2-11> Above <2-10>,
The second aspect of the obstacle position calculation method is used in the area (■) where formula <2-11> is calculated, and the line segment set within the area is the center line L2 drawn from the center O between the sensors to the area (■). It is. Part A indicates the inclination angle of this line segment with respect to the direction of travel of the car ('P>O).

The obstacle detection means (1) finally outputs the position coordinates (0, l, oY) of the obstacle calculated by the obstacle coordinate calculation means 3.

The storage means 4 for storing the installation position of the sensor and the shape value of the measurable area is composed of a memory for storing values such as wl, ρ, φ, A, II, i, and ll. These values are written in the memory in advance, and the storage means 4 outputs these values to the area determination means 2 and the obstacle coordinate calculation means 3.

The running course storage means 11 is comprised of a memory 11 in which parameters necessary for running the vehicle on a predetermined course are stored in advance. Parameters necessary for running on a course include the type of movement, course shape values, running speed, etc. Data is written into the memory 11 in advance before running the vehicle.

Position/azimuth measuring means ■1 is left and right independent measurement wheels 21R, 2
1L and independent counter circuits 22R and 22L that count the number of rotations of the respective measurement wheels 21R and 21L within a certain period of time.
and an arithmetic circuit 23 that calculates the current position and direction of the vehicle from the counted values of the counter circuits 22R and 22L.

The driving control means (1) is based on the position and orientation of the vehicle obtained from the position and orientation measuring means (1) and the parameter values of the course stored in the driving course storage means (2), based on the current state of the vehicle. An acceleration calculation circuit 31 that calculates the acceleration value of the left and right drive wheels necessary to travel at a given speed on a travel course, and integrates the acceleration value to calculate the rotational speed that is the actual drive amount of the left and right drive wheels. It consists of an integrating circuit 32.

The drive means (1) is a servo amplifier 4 that controls the rotation of the motor 41 according to the amount of drive given from the travel control means (2).
2, and motor 41. It consists of drive parts such as a clutch 43, gears 44, and drive wheels 45.

The obstacle information processing means (1) calculates the obstacle position coordinates in the coordinate system fixed to the unmanned vehicle given by the obstacle detection means (v1) based on the current position and orientation of the vehicle obtained from the position and orientation measuring means (1). An absolute position calculation circuit 61 converts the unmanned vehicle driving location into one common coordinate system that can be compared with the observation results of ■ Obstacle storage means by comparing the distance of the detected obstacle position;
Compares the obstacle position correction means 62 for correcting the obstacle position stored in the obstacle storage means 1 and the obstacle position stored in the obstacle storage means 1 with the course shape value stored in the running course storage means 11, It consists of an avoidance necessity determination circuit 63 that determines whether the obstacle exists on the driving course to be traveled in the future and whether an avoidance operation is necessary.

That is, the absolute position calculation circuit 61 is the obstacle detection means (1)
The obstacle location coordinates (0++, Ov) obtained from The coordinate position expressed in one common absolute position coordinate system (
ON-, 0y-), and consists of an X-coordinate calculation circuit that calculates and outputs 0□ and a y-coordinate calculation circuit that calculates and outputs O□. The X-coordinate calculation circuit and the y-coordinate calculation circuit each perform calculations shown in the following equations.

Oxm-0+ -cos θ -Oy -sin
θ + X ・ <2-12>Ova"Ow
−sin θ + Oy −cos θ +
y-<2-13> This operation can be realized using a trigonometric function generator, multiplier, and adder.

Further, the obstacle storage means (1) comprises a memory 71 for storing data on the position of the obstacle.

And, avoidance operation determining means ■1 is obstacle information processing means ■
When it is determined in step 1 that there is an obstacle that needs to be avoided, a deceleration instruction circuit 81 calculates the necessary deceleration and sends it to the travel control means 111+, and the obstacle position is stored in the obstacle storage means 1. A detour route that avoids collisions with obstacles based on the course data stored in the driving course storage means I and the current position and orientation of the vehicle obtained from the position and orientation measuring means (2).

(avoidance route) calculation means 82 for calculating the avoidance route; and driving course storage means ■1 as a course to be traveled on the avoidance route in the future.
It consists of a rewriting circuit 83 for driving course data to be stored in the memory.

When the unmanned vehicle is running, the position/direction measuring means ■1
Two independent counter circuits 22R, 22L count the number of rotations of the left and right independent side rails 21R, 21L within a certain period of time, and the calculation circuit 23 calculates the current position and direction of the vehicle from the counted values of the counter circuits 22R, 22L. calculate.

In the travel control means ■1, the acceleration calculation circuit 31
Based on the position and orientation of the vehicle obtained from the position and orientation measurement means (2) and the parameter values of the course stored in the travel course storage means 11, a given travel course is determined from the current state of the vehicle at a given speed. Find the acceleration values of the left and right drive wheels necessary for driving. The 81-minute circuit 32 integrates the acceleration value and outputs the rotational speed which is the actual drive amount of the left and right drive wheels.

In the driving means (1), the servo amplifier 42 controls the rotation of the motor 41 according to the amount of drive given from the travel control means mI, and the rotation of the motor 41 is transmitted to the driving wheels via the clutch 43 and gear 44, so that the unmanned drive the car.

Obstacle detection means (1) has the following functions. In the distance measuring means 1, first, an ultrasonic transmitting circuit 52 transmits ultrasonic waves from two ultrasonic sensors 51R and 51L attached to the left and right sides of the front of the vehicle, facing forward as shown in FIG. .

If an obstacle exists within the obstacle measurable area of each sensor shown in FIG. 6, reflected waves from the obstacle will return. After transmitting ultrasonic waves, the two ultrasonic sensors 51R and 51L function as ultrasonic receivers and receive the reflected waves. The distance measurement circuits 53R and 53L are connected to each sensor 5.
The delay time from the time of transmission to the time of reception of the ultrasonic waves received at 1R and 51L is measured, and the distance to the obstacle that reflected the ultrasonic waves is calculated for each of the left and right sides. When no obstacle exists within the measurable area, a distance value greater than or equal to the maximum detection distance l1 of the sensor is calculated.

Next, the area determining means 2 determines which distance measuring circuit 53R, left or right,
Depending on the distance fqd1, dl at which the obstacle is detected by the controller 53L, it is determined which small region the obstacle exists in or whether the obstacle does not exist.

If an obstacle exists in area ■ and does not exist in area ■, the output logic signal 1nv01 of comparator CI becomes 1,
The output logic signal of comparator C2 is 0. becomes 1, and the output logic signal Sl of the AND circuit A1 inputting it becomes 1, and it is determined that an obstacle exists in the area (2).

If an obstacle exists in area ■ and does not exist in area ■, the output logic signal 1nvo+ of comparator CI becomes 1,
The output logic signal 02 of the comparator C2 becomes 1, and the output logic signal S2 of the AND circuit A2 which manually inputs it becomes 1, and it is determined that an obstacle exists in the area (2).

If the obstacle exists only in the area ■, the output logic signal 1nvo+ of the comparator C1 becomes 1, the output logic signal 1nvot of the comparator C2 becomes 1, and the output logic signal of the AND circuit A4 inputting it becomes 1. S4 becomes l. As a result, comparators C4 and C5 are activated, and their respective output logic signals 1nvoa and OS become 1. The output logic signal S of 6 becomes 1 to the AND circuit that inputs it.

When output logic signal S6 is 1, comparator C7 is activated;
Output logic signal ○! , and it is determined that there is an obstacle in area ■.

In the case where the obstacle exists only in the area ■ and the case where the obstacle exists only in the area ■, it is similarly determined that the obstacle exists in the area ■ and the area ■, respectively.

Also, when multiple obstacles exist in different areas,
If a large obstacle exists over several areas,
The area where the obstacle exists closest to the unmanned vehicle is selected as the area where the obstacle exists. An example of determination of this area is shown in FIG. In FIG. 7, the X mark indicates the position where the obstacle actually exists, and the diagonal line indicates the selected area.

The region determining means 2 outputs a logical signal representing the region where the determined obstacle exists or a logical signal representing the absence of the obstacle.

One of the five coordinate calculation circuits P1 to P5 corresponding to each area of the obstacle coordinate calculation means 3 is activated by a logic signal representing the determination area outputted from the area determination means 2. That is, for example, when it is determined that an obstacle exists in area (2), the coordinate calculation circuit PI is activated. Within the coordinate calculation circuits P1 to P5, X coordinate calculation circuits P11 to R51 that calculate the X coordinate of the obstacle and Y coordinate calculation circuits P12 to P52 that calculate the Y coordinate of the obstacle are activated. Based on the distance values from the distance measurement circuits 53R and 53L, the obstacle position coordinates in the coordinate system fixed to the unmanned vehicle are calculated. When calculating the position coordinates of an obstacle in areas ■, ■, ■, ■, it is approximated that the obstacle exists at the detected distance position on the center lines L1, L2, R1, R2 assumed within the area. . The relationship between the actual obstacle position and the calculated obstacle position in this case is shown in FIG.

The x mark in Figure 8 indicates the position where an obstacle actually exists, and the
Marks indicate approximated obstacle positions.

In addition, when calculating the position coordinates of an obstacle in area (3), the position coordinates of the obstacle are calculated by finding the intersection of circles whose centers are the installation positions of each sensor and whose radii are each distance value. Therefore, there is no error due to approximation like in other areas.

Also, although an obstacle is actually not a point but has a certain size, the distance value obtained from the distance measurement means is the distance on the point closest to the obstacle sensor, so in the end there is only one point at that distance. The point closest to the obstacle will be output as the position of the obstacle.

The storage means 4 for storing the installation position of the distance measuring means and the shape value of the measurable area includes the installation position of the sensors 51R and 51L necessary for the processing of the area determination means 3 and the obstacle coordinate calculation means 3, and the shape value of the measurable area. be memorized in advance.

As a result of the above processing, the obstacle detecting means (1) outputs the position coordinates of one point closest to the vehicle of the obstacle within the obstacle measurable area based on only one measurement result of the distance measuring means.

In the obstacle information means (1), when an obstacle exists, the absolute position calculation circuit 61 calculates the obstacle position (OX
, Or) and the current position and orientation of the vehicle obtained from the position and orientation measuring means (2), a conversion calculation is performed into one common coordinate system at the location where the unmanned vehicle is traveling, which can be compared with previous observation results. The obstacle position correction means 62 compares the distance of the previously detected obstacle position stored in the memory 71 of the obstacle storage means 1 with the distance of the currently detected obstacle position whose coordinates have been transformed. 0 Avoidance necessity determination circuit 6 for correcting the obstacle position stored in the memory 71 of object storage means 1
3 is the corrected obstacle position M stored in the memory 71 of obstacle storage means 1 (input from the obstacle position correction means 62)
The obstacle is compared with the shape value of the course stored in the driving course storage means (1), and it is determined whether the obstacle exists on the driving course to be traveled in the future and an avoidance action is required.

In the avoidance operation determining means (2), when the obstacle information processing means (1) determines the existence of an obstacle that requires avoidance, the deceleration instruction circuit 81 calculates and sends the deceleration required to the travel control means (1). do. The travel control means (1) decelerates the unmanned vehicle according to this deceleration. - On the other hand, the avoidance route calculation means 82 uses the obstacle position stored in the obstacle storage means ■1, the course data stored in the driving course storage means ■1, and the current position of the vehicle obtained from the position and orientation measuring means ■1. Based on the direction, a detour (avoidance route) to avoid collision with an obstacle is calculated. The driving course data rewriting circuit 83 uses the avoidance route outputted by the avoidance route calculation means 82 as the next course to be traveled as a driving course storage means! 1
to be memorized. Once the unmanned vehicle has decelerated in response to the deceleration instruction, the avoidance route is stored in the travel course storage means ■1, and the unmanned vehicle then uses the avoidance route as the target travel course and performs the same travel control as normal to follow that course. conduct.

After completing the avoidance route, return to the original driving course,
The avoidance action ends.

The device of this embodiment having the above-mentioned function has an obstacle detection means v
Since the obstacle information output from 1 is given in the form of the coordinates of one point of the closest part of the obstacle to the unmanned vehicle within the measurable 8M area, it is possible to avoid obstacles that are unnecessary for simple avoidance operations. The complex external shape information of the vehicle is compressed and only the necessary information is extracted in a clean form, making it easier to use this information to perform processes such as determining the operation of the unmanned vehicle. Furthermore, since the process is easy, there is also the advantage that the equipment used for this process can be small-scale.

In addition, the obstacle information obtained with one detection operation is the coordinates of one point closest to the unmanned vehicle, but the detection operation is repeated while moving and compared with the previously obtained obstacle position information. Therefore, if the same point of an obstacle is detected several times, the accuracy of the detection position can be improved, and if different points of the obstacle are detected, the distribution situation of the obstacle can be grasped, and the accuracy of the detection position can be improved. Accurate obstacle information can be obtained.

As a result, flexible avoidance actions can be performed in response to various obstacle situations.

In addition, by using two fixed ultrasonic sensors, obstacles can be detected in the required area with a short process of one distance measurement operation for each sensor, so obstacles are discovered quickly and avoidance actions are delayed until collision occurs. There is little risk of

Furthermore, in order to calculate the position coordinates of the obstacle, the entire obstacle measurable area is divided into small areas, and the area determination means 2 is installed to limit the range of the obstacle within the small area, thereby calculating the position coordinates. It is possible to easily control the error in the position coordinates and to suppress the error in the position coordinates within a certain range.

Furthermore, the obstacle coordinate calculation means for calculating the position coordinates of obstacles assumes a line segment within the area in areas ■, ■, ■, ■, and calculates the given distance value on this line segment. Using the method of calculating the corresponding coordinates, in the example, we adopted the center line drawn from the sensor to the area as this line segment, so by using the center line as this line segment, the error in calculating the obstacle position can be minimized. can be suppressed to Also, in area ■, 2
We used a method to calculate the position coordinates of the obstacle by using the distance values from two sensors and finding the intersection of circles with each sensor installation position as the center and each distance II lfi as the radius. If both sensors capture the same point on the obstacle, the coordinates of the obstacle can be calculated accurately without approximation.

If this embodiment device is applied to a collision avoidance device for an unmanned vehicle,
Two ultrasonic sensors detect obstacles in front of the vehicle in the area the vehicle will pass, and if an obstacle is present, the vehicle will take a detour to avoid collision with the obstacle. This allows collisions with obstacles to be avoided.

In the above embodiment, an ultrasonic sensor is used as the sensor of the distance measuring means, but other sensors may be used. Furthermore, although the hardware configuration of the obstacle coordinate calculation means 3 and the area determination means 2 using a dedicated calculation circuit is shown, it is also possible to use a microcomputer and use a program for that purpose, and the program can be easily changed. For this reason, it is practically better to use a microcomputer.

[Brief explanation of drawings]

FIG. 1, FIG. 2, and FIG. 3 are explanatory diagrams showing the prior art;
Figure 4 is a schematic configuration diagram showing the concept of the present invention, Figure 5 (1)
is a system diagram showing the entire embodiment of the present invention, FIG. 5 (2)
is the circuit diagram of the area determination means in Figure 5+11, Figure 5 (3)
5+11 is a circuit diagram of the obstacle coordinate calculation means, and FIG. 6 is the installation position of the sensor of the embodiment, the shape value of the measurable area,
FIG. 7 is a plan view showing an example of the operation of the area determination means of the embodiment; FIG. 8 is a plan view showing an example of the operation of the obstacle coordinate calculation means of the embodiment. It is a diagram. ■1...Traveling course storage means, ■1...Position and orientation measuring means, ■1...Travel control means, ■1...Drive means, ■,...Obstacle detection means, ■1. ... Obstacle information processing means, ■1 ... Obstacle storage means, ■1 ... Avoidance operation setting means.

Claims (4)

[Claims] (1) Obstacle detection devices for unmanned vehicles that move autonomously along a preset driving course detect obstacles that exist within a measurable area that is smaller than the range that requires obstacle detection when the unmanned vehicle is running. at least two sensors that measure the distance of the sensor, the measurable area partially overlaps so that the entire measurable area covers the necessary area, and is smaller than the measurable area and measurable by one sensor. a distance measuring means in which the sensor is installed so as to be able to divide the entire measurable area into a small area and a small area that is smaller than the measurable area and measurable by at least two sensors; A storage means for storing in advance a shape value of a measurable area; and a storage means for determining which sensor has detected an obstacle when the sensor detects the obstacle, and determining the installation position of the sensor stored in the storage means; an area determining means for determining a small area in which an obstacle exists based on a shape value of the measurable area; and a distance value measured by the sensor, the installation position of the sensor, and the shape of the measurable area stored in the storage means. and an obstacle coordinate calculating means for estimating the position of the obstacle within the small area determined by the area determining means based on the value and calculating the obstacle position coordinate in a coordinate system fixed to the unmanned vehicle. An obstacle detection device for unmanned vehicles featuring: (2) The area determination means determines which sensor has detected the obstacle when the sensor detects the obstacle, and also determines the installation position of the sensor and the shape of the measurable area stored in the storage means. An obstacle detection device for an unmanned vehicle according to claim 1, which determines a small area where an obstacle exists based on the distance value and the distance value output from the sensor that has detected the obstacle. (3) The obstacle coordinate calculation means estimates the position of the obstacle within the small area determined by the area determination means based on the sensor installation position and the shape value of the measurable area stored in the storage means. At the same time, a representative line is assumed within the small area, and the obstacle in the coordinate system fixed to the unmanned vehicle is calculated by calculating the coordinates of the point corresponding to the distance value output from the sensor that detects the obstacle on the representative line. An obstacle detection device for an unmanned vehicle according to claim (1), which calculates object position coordinates. (4) When the area determination means determines that an obstacle exists in a portion where the measurable areas partially overlap, the obstacle coordinate calculation means outputs a distance value output from a sensor that has detected the obstacle. and the installation position of each sensor stored in the storage means, the obstacle position is calculated by calculating the coordinates of the intersection of a circle whose center is the installation position of each sensor and whose radius is each distance value. An obstacle detection device for an unmanned vehicle according to claim (1), which calculates coordinates.