Obstacle ranging method for intelligent vehicles

Abstract

The present invention provides an obstacle ranging method for an intelligent vehicle, comprising the following steps: a laser radar detects an obstacle, wherein a plurality of obstacle avoidance detection channels are preset in the laser radar, and the obstacle avoidance detection channel currently detecting the obstacle is set as the obstacle avoidance detection channel SpdIndex; the obstacle avoidance detection channel is provided with longitudinal detection areas LXF, LXM, and LXN, and when the obstacle is located outside the LXF area and within the LXF, LXM, and LXN areas, a corresponding feedback signal is transmitted to the intelligent vehicle; when the feedback signal of the laser radar changes, the change in the feedback signal is judged to obtain the obstacle distance D, and the obstacle avoidance detection channel SpdIndex is updated; and a target speed v is calculated according to the obstacle distance D. target ; Intelligent vehicle is adjusted to target speed v target The obstacle distance measurement method of the present invention integrates and calculates multiple IO signals of a low-precision laser radar to obtain high-precision distance information.

Description

Technical Field

[0001] The present invention relates to the field of AGV obstacle detection technology, and in particular to an obstacle ranging method for an intelligent vehicle. Background Art

[0002] The main content of AGV's obstacle detection technology is to let AGV know the distance of obstacles around it, so that AGV can execute corresponding obstacle strategies in advance. The existing AGV obstacle detection method is to equip AGV with laser radar, use the laser radar to detect obstacles and then feedback distance information, and AGV then performs obstacle operation according to the distance information. The higher the detection accuracy of the laser radar, the more accurate the feedback distance information. However, due to the high cost of high-precision laser radar (which can feedback more than dozens of IO signals), it is not suitable for AGV projects with low project budgets. In order to reduce costs, the AGV of this type of AGV project is equipped with low-precision laser radar that can only feedback IO signals of several channels. Since the low-precision radar cannot accurately feedback distance information, the effectiveness of AGV in performing obstacle operations is reduced. Summary of the Invention

[0003] The purpose of the present invention is to overcome the shortcomings of the existing technology and provide an obstacle ranging method for intelligent vehicles, which integrates and calculates multiple IO signals of low-precision lidar to obtain high-precision distance information.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions:

[0005] The obstacle ranging method of the intelligent vehicle includes the following steps:

[0006] The laser radar installed on the intelligent vehicle detects obstacles. The laser radar has several preset obstacle avoidance detection channels. The obstacle avoidance detection channel currently detecting the obstacle is set as the obstacle avoidance detection channel SpdIndex;

[0007] Each obstacle avoidance detection channel SpdIndex is configured with three longitudinal detection zones: LXF, LXM, and LXN. X represents the obstacle avoidance detection channel number. When an obstacle is outside the LXF zone, or within the LXF, LXM, and LXN zones, corresponding feedback signals are transmitted to the intelligent vehicle. The longitudinal detection zone of the previous obstacle avoidance detection channel is the sum of the lengths of the LXM and LXN detection zones of the subsequent obstacle avoidance detection channel.

[0008] When the feedback signal of the lidar changes, the change of the feedback signal is judged to obtain the obstacle distance D and update the obstacle avoidance detection channel SpdIndex;

[0009] Calculate the target speed v according to the obstacle distance D target ;

[0010] Smart vehicles are adjusted to move at target speed v target Move.

[0011] Compared with the existing technology, the obstacle ranging method of the intelligent vehicle of the present invention uses a low-precision laser radar that is inexpensive and can only feedback IO signals of several channels. By switching the obstacle avoidance detection channel in real time and integrating multiple IO signals of the low-precision laser radar, the obstacle distance D is calculated, thereby realizing that as the AGV gradually approaches the obstacle, the obstacle distance D will gradually decrease, and the target speed v target The technical effect of gradually reducing (slowing down) accordingly achieves the obstacle detection effect that can be achieved by expensive high-precision lidar.

[0012] Preferably, when the obstacle is outside the LXF area, the laser radar feedback signal is S; when the obstacle is in the LXF area, the laser radar feedback signal is F; when the obstacle is in the LXM area, the laser radar feedback signal is M; when the obstacle is in the LXN area, the laser radar feedback signal is N;

[0013] When the feedback signal from the LiDAR changes, the following two judgments are performed simultaneously:

[0014] a. Whether the obstacle avoidance detection channel number determined in the current working cycle is the same as the obstacle avoidance detection channel number determined in the previous working cycle;

[0015] b. Determine whether the feedback signal has changed;

[0016] If item a is judged to be the same, and the feedback signal of item b switches from S to F, the longitudinal length corresponding to LXF of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D;

[0017] If item a is judged to be the same, and the feedback signal of item b switches from M to F, the longitudinal length corresponding to LXM of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D;

[0018] If item a is judged to be the same, and the feedback signal of item b switches from F to S, the longitudinal length corresponding to LXF of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D, and the obstacle avoidance detection channel (SpdIndex+1) is used as the new obstacle avoidance detection channel SpdIndex after correction;

[0019] If item a is judged to be the same, and the feedback signal of item b switches from F to M, the longitudinal length corresponding to LXM of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D. After correction, the obstacle avoidance detection channel (SpdIndex-1) is used as the new obstacle avoidance detection channel SpdIndex;

[0020] If item a is judged to be the same, and the feedback signal of item b switches from M to N, the longitudinal length corresponding to LXN of the current obstacle avoidance detection channel spdIndex is used as the obstacle distance D;

[0021] If a is judged to be different and the feedback signal switches from M to F, the longitudinal length corresponding to LXF of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D;

[0022] If item a is judged to be different and the feedback signal of item b switches from M to N, the longitudinal length corresponding to LXN of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D.

[0023] The above judgment and classification analysis are performed through the feedback signal of the lidar, so that the obstacle distance D can be calculated in real time according to the switching of the feedback signal of the obstacle avoidance detection channel, thereby improving the accuracy of the obstacle distance D and the effectiveness of obstacle avoidance.

[0024] Preferably, if a situation that does not meet the above 7 judgment results occurs, the obstacle distance D is calculated using the encoding device increment.

[0025] In order to cope with the problems that may arise when an intelligent vehicle travels on different terrains during actual production, the present invention is also provided with a method for calculating the obstacle distance D by increments of the encoding device, thereby improving the reliability of the present invention.

[0026] Preferably, the formula for calculating the obstacle distance D based on the incremental calculation of the encoder device of the driving wheel is:

[0027] Current speed = encoder increment ÷ (number of pulses × reducer gear ratio) × (π × wheel diameter) ÷ encoder reporting cycle;

[0028] Obstacle distance D = obstacle distance D in the previous cycle + current speed × interpolation time interval, where the interpolation time interval is 0.01-100 milliseconds.

[0029] Preferably, the laser radar is provided with x obstacle avoidance detection channels, and the x obstacle avoidance detection channels are numbered 1 to n respectively;

[0030] Generate a mapping set R for each obstacle avoidance detection channel SpdIndex and the corresponding preset speed range;

[0031] According to the mapping set R, obtain the predicted obstacle avoidance detection channel SpdIndexText corresponding to the current moving speed of the intelligent vehicle;

[0032] After executing items a and b, determine the size of the obstacle avoidance detection channel number corresponding to the predicted obstacle avoidance detection channel SpdIndexText and the obstacle avoidance detection channel number corresponding to the obstacle avoidance detection channel SpdIndex, and select the obstacle avoidance detection channel with the smaller number as the new obstacle avoidance detection channel SpdIndex.

[0033] Since the longitudinal detection area of ​​the previous obstacle avoidance detection channel is the sum of the lengths of the LXM and LXN detection areas of the subsequent obstacle avoidance detection channel (the longitudinal detection area of ​​the subsequent obstacle avoidance detection channel is larger than that of the previous obstacle avoidance detection channel, and the previous obstacle avoidance detection channel corresponds to a smaller number of lidar obstacle avoidance channel than that of the subsequent obstacle avoidance detection channel), the above setting method enables the intelligent vehicle to always use the obstacle avoidance detection channel SpdIndex with a smaller number as the key obstacle avoidance detection channel.

[0034] Preferably, the LXN detection area detection ranges of the various obstacle avoidance detection channels are the same.

[0035] Preferably, according to the formula Calculate the target speed v target , where a dec_max The preset maximum braking deceleration.

[0036] Preferably, the target speed v is obtained target Then, calculate the motor speed parameters according to the following formula:

[0037]

[0038] Where Motor_Dac is the motor speed per minute, RearWheel_Diameter is the drive wheel diameter, and RearWheeLxMotor_TransmissionRatio is the drive wheel reduction ratio;

[0039] The motor speed parameters are transmitted to the travel driver of the intelligent vehicle, and the travel driver controls the travel motor speed to the target speed according to the motor speed parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is a flow chart of the present invention;

[0041] Figure 2 It is a flow chart to obtain the obstacle distance D;

[0042] Figure 3 Schematic diagram showing the LXF, LXM, and LXN detection areas of the eight obstacle avoidance detection channels of the lidar. DETAILED DESCRIPTION

[0043] The following describes the embodiments of the present invention with reference to the accompanying drawings:

[0044] See also Figures 1 to 2 The obstacle distance measurement method for the intelligent vehicle of this embodiment includes the following steps:

[0045] The laser radar installed on the intelligent vehicle detects obstacles. The laser radar has several preset obstacle avoidance detection channels. The obstacle avoidance detection channel currently detecting the obstacle is set as the obstacle avoidance detection channel SpdIndex;

[0046] Each obstacle avoidance detection channel SpdIndex is configured with three longitudinal detection zones: LXF, LXM, and LXN. X represents the obstacle avoidance detection channel number. When an obstacle is outside the LXF zone, or within the LXF, LXM, and LXN zones, corresponding feedback signals are transmitted to the intelligent vehicle. The longitudinal detection zone of the previous obstacle avoidance detection channel is the sum of the lengths of the LXM and LXN detection zones of the subsequent obstacle avoidance detection channel.

[0047] When the feedback signal of the lidar changes, the change of the feedback signal is judged to obtain the obstacle distance D and update the obstacle avoidance detection channel SpdIndex;

[0048] Calculate the target speed v according to the obstacle distance D target ;

[0049] Smart vehicles are adjusted to move at target speed v target Move.

[0050] Specifically, the laser radar is arranged in front of the moving direction of the intelligent vehicle.

[0051] Specifically, the longitudinal distances of the three longitudinal detection areas LXF, LXM, and LXN of each of the x obstacle avoidance detection channels are preset parameters, that is, the longitudinal distances of the three longitudinal detection areas of the obstacle avoidance detection channels are known.

[0052] Preferably, when the obstacle is outside the LXF area, the obstacle avoidance detection channel feedback signal is S; when the obstacle is in the LXF area, the obstacle avoidance detection channel feedback signal is F; when the obstacle is in the LXM area, the obstacle avoidance detection channel feedback signal is M; when the obstacle is in the LXN area, the obstacle avoidance detection channel feedback signal is N;

[0053] Specifically, the feedback signal S represents security, the feedback signal F represents far, the feedback signal M represents middle, and the feedback signal N represents near.

[0054] When the feedback signal from the LiDAR changes, the following two judgments are performed simultaneously:

[0055] a. Whether the obstacle avoidance detection channel number determined in the current working cycle is the same as the obstacle avoidance detection channel number determined in the previous working cycle; (Since this method operates cyclically, there are current and previous working cycles, and each working cycle calculates and determines the primary obstacle avoidance detection channel for focused obstacle detection).

[0056] b. Determine whether the feedback signal has changed;

[0057] If item a is judged to be the same, and the feedback signal of item b switches from S to F, the longitudinal length corresponding to LXF of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D;

[0058] If item a is judged to be the same, and the feedback signal of item b switches from M to F, the longitudinal length corresponding to LXM of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D;

[0059] If item a is judged to be the same, and the feedback signal of item b switches from F to S, the longitudinal length corresponding to LXF of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D, and the obstacle avoidance detection channel (SpdIndex+1) is used as the new obstacle avoidance detection channel SpdIndex after correction;

[0060] If item a is judged to be the same, and the feedback signal of item b switches from F to M, the longitudinal length corresponding to LXM of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D. After correction, the obstacle avoidance detection channel (SpdIndex-1) is used as the new obstacle avoidance detection channel SpdIndex;

[0061] If item a is judged to be the same, and the feedback signal of item b switches from M to N, the longitudinal length corresponding to LXN of the current obstacle avoidance detection channel spdIndex is used as the obstacle distance D;

[0062] If a is judged to be different and the feedback signal switches from M to F, the longitudinal length corresponding to LXF of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D;

[0063] If item a is judged to be different and the feedback signal of item b switches from M to N, the longitudinal length corresponding to LXN of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D.

[0064] The above judgment and classification analysis are performed through the feedback signal of the lidar, so that the obstacle distance D can be calculated in real time according to the switching of the feedback signal of the obstacle avoidance detection channel, thereby improving the accuracy of the obstacle distance D and the effectiveness of obstacle avoidance.

[0065] Preferably, if a situation that does not meet the above 7 judgment results occurs, the obstacle distance D is calculated using the encoding device increment.

[0066] In order to cope with the problems that may arise when an intelligent vehicle travels on different terrains during actual production, the present invention is also provided with a method for calculating the obstacle distance D by increments of the encoding device, thereby improving the reliability of the present invention.

[0067] Preferably, the formula for calculating the obstacle distance D based on the incremental calculation of the encoder device of the driving wheel is:

[0068] Current speed = encoder increment ÷ (number of pulses × reducer gear ratio) × (π × wheel diameter) ÷ encoder reporting cycle;

[0069] Obstacle distance D = obstacle distance D in the previous cycle + current speed × interpolation time interval, where the interpolation time interval is 0.01-100 milliseconds.

[0070] Preferably, the laser radar is provided with x obstacle avoidance detection channels, and the x obstacle avoidance detection channels are numbered 1 to n respectively;

[0071] Generate a mapping set R for each obstacle avoidance detection channel SpdIndex and the corresponding preset speed range;

[0072] According to the mapping set R, obtain the predicted obstacle avoidance detection channel SpdIndexText corresponding to the current moving speed of the intelligent vehicle;

[0073] After executing the judgments in items a and b, determine the size of the obstacle avoidance detection channel number corresponding to the predicted obstacle avoidance detection channel SpdIndexText and the obstacle avoidance detection channel corresponding to the obstacle avoidance detection channel SpdIndex (the obstacle avoidance detection channel SpdIndex here is the obstacle avoidance detection channel number SpdIndex obtained by the judgments in items a and b above), and select the obstacle avoidance detection channel with the smaller number as the new obstacle avoidance detection channel SpdIndex (the lidar will switch the obstacle avoidance detection channel with the smaller number as the new channel SpdIndex).

[0074] Since the longitudinal detection area of ​​the previous obstacle avoidance detection channel is the sum of the lengths of the LXM and LXN detection areas of the subsequent obstacle avoidance detection channel (the longitudinal detection area of ​​the subsequent obstacle avoidance detection channel is larger than that of the previous obstacle avoidance detection channel, the previous obstacle avoidance detection channel can detect closer obstacles, and the lidar obstacle avoidance channel number corresponding to the previous obstacle avoidance detection channel is smaller than the lidar obstacle avoidance channel number corresponding to the subsequent obstacle avoidance detection channel), the above setting method can enable the intelligent vehicle to always use the lidar obstacle avoidance detection channel SpdIndex with a smaller number as the key obstacle avoidance detection channel.

[0075] Preferably, the LXN detection area detection ranges of the various obstacle avoidance detection channels are the same.

[0076] The following distance description applies to the application example of the laser radar of this embodiment:

[0077] This application example shows the eight obstacle avoidance detection channels of the laser radar. The longitudinal detection areas of the eight obstacle avoidance detection channels are as follows: Figure 3 As shown in the figure, the longitudinal detection areas of the 8 obstacle avoidance detection channels are numbered as follows:

[0078] Obstacle avoidance detection channel No. 1: L1F, L1M and L1N;

[0079] Obstacle avoidance detection channel 2: L2F, L2M and L2N;

[0080] Obstacle avoidance detection channel 3: L3F, L3M and L3N;

[0081] Obstacle avoidance detection channel 4: L4F, L4M and L4N;

[0082] Obstacle avoidance detection channel No. 5: L5F, L5M and L5N;

[0083] Obstacle avoidance detection channel No. 6: L6F, L6M and L6N;

[0084] Obstacle avoidance detection channel No. 7: L7F, L7M and L7N;

[0085] Obstacle avoidance detection channel 8: L8F, L8M and L8N;

[0086] The case of the mapping set R of 8 obstacle avoidance detection channels:

[0087] Table 1 Mapping set R

[0088] Obstacle avoidance detection channel number Preset speed range (unit: m / s) 1 0～0.05 2 0.05～0.15 3 0.15～0.3 4 0.3～0.6 5 0.6～0.9 6 0.9～1.2 7 1.2～1.4 8 1.4～1.6

[0089] Preferably, according to the formula Calculate the target speed v target , where a dec_max The preset maximum braking deceleration.

[0090] Preferably, the target speed v is obtained target Then, calculate the motor speed parameters according to the following formula:

[0091]

[0092] Where Motor_Dac is the motor speed per minute, RearWheel_Diameter is the drive wheel diameter, and RearWheeLxMotor_TransmissionRatio is the drive wheel reduction ratio;

[0093] The motor speed parameters are transmitted to the travel driver of the intelligent vehicle, and the travel driver controls the travel motor speed to the target speed according to the motor speed parameters.

[0094] Compared with the existing technology, the obstacle ranging method of the intelligent vehicle of the present invention uses a low-precision laser radar that is inexpensive and can only feedback IO signals of several channels. By switching the obstacle avoidance detection channel in real time and integrating multiple IO signals of the low-precision laser radar, the obstacle distance D is calculated, thereby realizing that as the AGV gradually approaches the obstacle, the obstacle distance D will gradually decrease, and the target speed v target The technical effect of gradually reducing (slowing down) accordingly achieves the obstacle detection effect that can be achieved by expensive high-precision lidar.

[0095] Based on the disclosure and teachings of the above description, those skilled in the art may also make changes and modifications to the above embodiments. Therefore, the present invention is not limited to the specific embodiments disclosed and described above, and modifications and variations of the present invention should also fall within the scope of protection of the claims of the present invention. In addition, although certain specific terms are used in this description, these terms are only for convenience of description and do not constitute any limitation to the present invention.

Claims

1. An obstacle ranging method for an intelligent vehicle includes the following steps: The laser radar installed on the intelligent vehicle detects obstacles. The laser radar has several preset obstacle avoidance detection channels. The obstacle avoidance detection channel currently detecting the obstacle is set as the obstacle avoidance detection channel SpdIndex; Each obstacle avoidance detection channel is equipped with three longitudinal detection areas: LXF, LXM, and LXN. X represents the obstacle avoidance detection channel number. When an obstacle is outside the LXF area, or within the LXF, LXM, and LXN areas, corresponding feedback signals are transmitted to the intelligent vehicle. The longitudinal detection area of ​​the previous obstacle avoidance detection channel is the sum of the lengths of the LXM and LXN detection areas of the subsequent obstacle avoidance detection channel. When the feedback signal of the lidar changes, the change of the feedback signal is judged to obtain the obstacle distance D and update the obstacle avoidance detection channel SpdIndex; Calculate the target speed v according to the obstacle distance D target ; Smart vehicles are adjusted to move at target speed v target to move; When the obstacle is outside the LXF area, the laser radar feedback signal is S; when the obstacle is in the LXF area, the laser radar feedback signal is F; when the obstacle is in the LXM area, the laser radar feedback signal is M; when the obstacle is in the LXN area, the laser radar feedback signal is N; When the feedback signal from the LiDAR changes, the following two judgments are performed simultaneously: a. Whether the obstacle avoidance detection channel number determined in the current working cycle is the same as the obstacle avoidance detection channel number determined in the previous working cycle; b. Determine whether the feedback signal has changed; If item a is judged to be the same, and the feedback signal of item b switches from S to F, the longitudinal length corresponding to LXF of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D; If item a is judged to be the same, and the feedback signal of item b switches from M to F, the longitudinal length corresponding to LXM of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D; If item a is judged to be the same, and the feedback signal of item b switches from F to S, the longitudinal length corresponding to LXF of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D, and the obstacle avoidance detection channel (SpdIndex+1) is used as the new obstacle avoidance detection channel SpdIndex after correction; If item a is judged to be the same, and the feedback signal of item b switches from F to M, the longitudinal length corresponding to LXM of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D. After correction, the obstacle avoidance detection channel (SpdIndex-1) is used as the new obstacle avoidance detection channel SpdIndex; If item a is judged to be the same, and the feedback signal of item b switches from M to N, the longitudinal length corresponding to LXN of the current obstacle avoidance detection channel spdIndex is used as the obstacle distance D; If a is judged to be different and the feedback signal switches from M to F, the longitudinal length corresponding to LXF of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D; If item a is judged to be different and the feedback signal of item b switches from M to N, the longitudinal length corresponding to LXN of the current obstacle avoidance detection channel SpdIndex is used as the obstacle distance D; The laser radar is equipped with x obstacle avoidance detection channels, which are numbered from 1 to n. Generate a mapping set R for each obstacle avoidance detection channel SpdIndex and the corresponding preset speed range; According to the mapping set R, obtain the predicted obstacle avoidance detection channel SpdIndexText corresponding to the current moving speed of the intelligent vehicle; After executing items a and b, determine the size of the obstacle avoidance detection channel number corresponding to the predicted obstacle avoidance detection channel SpdIndexText and the obstacle avoidance detection channel number corresponding to the obstacle avoidance detection channel SpdIndex, and select the obstacle avoidance detection channel with the smaller number as the new obstacle avoidance detection channel SpdIndex.

2. The obstacle distance measurement method for an intelligent vehicle according to claim 1, characterized in that: If a situation that does not meet the above 7 judgment results occurs, the obstacle distance D is calculated based on the incremental calculation of the encoding device of the driving wheel.

3. The obstacle distance measurement method for an intelligent vehicle according to claim 2, characterized in that: The formula for calculating the obstacle distance D based on the incremental calculation of the encoder device of the driving wheel is: Current speed = encoder increment ÷ (number of pulses × reducer gear ratio) × (π × wheel diameter) ÷ encoder reporting cycle; Obstacle distance D = obstacle distance D in the previous cycle + current speed × interpolation time interval, where the interpolation time interval is 0.01-100 milliseconds.

4. The obstacle distance measurement method for an intelligent vehicle according to claim 1, characterized in that: The LXN detection area detection range of each obstacle avoidance detection channel is the same.

5. The obstacle distance measurement method for an intelligent vehicle according to claim 1, characterized in that: According to the formula Calculate the target speed v target , where a dec_max The preset maximum braking deceleration.

6. The obstacle distance measurement method for an intelligent vehicle according to claim 1, characterized in that: Get the target speed v target Then, calculate the motor speed parameters according to the following formula: Where Motor_Dac is the motor speed per minute, RearWheel_Diameter is the drive wheel diameter, and RearWheeLxMotor_TransmissionRatio is the drive wheel reduction ratio; The motor speed parameters are transmitted to the travel driver of the intelligent vehicle, and the travel driver controls the travel motor speed to the target speed according to the motor speed parameters.

Images