CN113277046B - A depth-fixing control method for a manta ray-like underwater vehicle based on the centroid and caudal fin - Google Patents

Abstract

The invention relates to a method for controlling the depth setting of an imitated manta ray underwater vehicle based on a centroid and a tail fin, which aims at propelling the bionic underwater vehicle by a pectoral fin with strong model uncertainty, divides a depth setting task into three stages by adopting segmentation, and implements online correction on PID (proportion integration differentiation) coefficients by combining a fuzzy controller to realize the adjustment of the centroid structure and the tail fin so as to complete the task of swimming at the depth. In the depth setting process, a pectoral fin mechanism of the underwater vehicle is driven by a steering engine and always keeps a flapping state, and the steering engine outputs a constant amplitude value and a constant phase difference to generate thrust along an X axis of a carrier coordinate system. The pitching moment is generated by changing the back and forth movement of the metamorphic heart block and the up and down deflection of the tail fin, so that the depth fixing effect is achieved.

Description

A depth-fixing control method for a manta ray-like underwater vehicle based on the centroid and caudal fin

technical field

The invention belongs to the field of motion control of underwater vehicles, and relates to a depth-fixing control method of a manta ray-like underwater vehicle based on a center of mass and a tail fin. method.

Background technique

The manta-like underwater vehicle is a new type of bionic underwater vehicle. It adopts the pectoral fin propulsion mode and has significant advantages such as high efficiency, high maneuverability, high stability, and low disturbance. It can achieve excellent in-situ observation and adapt to complex At the same time, it also has good biological affinity, which can be used in the fields of coral starfish disaster monitoring and fish condition monitoring in marine pastures, and has broad application prospects and theoretical research value.

The model of the pectoral fin propulsion bionic underwater vehicle is very complex, so far there is no relatively accurate model of the pectoral fin oscillating propulsion fish, which limits the application of traditional control methods in the bionic vehicle. At present, the commonly used depth-fixing control methods for this type of aircraft include fuzzy control algorithm, PID control algorithm, etc., but the simple information processing of the fuzzy control algorithm will lead to the reduction of the system control accuracy and the deterioration of the dynamic performance; the parameters in the PID control algorithm are fixed The value does not change with the system and does not have self-tuning characteristics. In the published literature, there is little mention of the specific method of depth-determination control of the pectoral fin bionic underwater vehicle, such as the patent: a fish-shaped bionic underwater robot and its control method [P].CN111284663B and patent: bionic manta ray robot[P]. P]. CN112093018A.

SUMMARY OF THE INVENTION

technical problem to be solved

In order to avoid the deficiencies of the prior art, the present invention proposes a manta ray-like underwater vehicle depth determination control method based on the center of mass and caudal fin, which is a manta ray-like underwater vehicle based on the cooperative control of the center of mass mechanism and the caudal fin. depth control method.

Technical solutions

A method for controlling the depth of an imitation manta ray underwater vehicle based on the center of mass and the tail fin, characterized in that the steps are as follows:

Step 1. Calculate the depth deviation and the depth deviation rate of change: the current depth of the underwater vehicle obtained through the depth sensor is y, and the downward direction is positive; the reference depth set by the task is y _d , then the depth deviation Δy is:

Δy=yy _d

Taking the derivation of the depth deviation, the depth deviation change rate v is obtained as:

Among them: t is the depth sensor information update cycle of the underwater vehicle;

Step 2. Divide the depth setting task into the following three working conditions with the depth deviation as input:

(1) Rapid dive/ascent: when Δy≥a, that is, the current depth is far from the target depth, and the underwater vehicle is in the rapid dive/ascent stage, the large fuzzy PID proportional coefficient composed of Table 1 and Table 3 is used to set scope;

(2) Start depth determination: when a≥Δy≥b, the current depth is close to the target depth, the underwater vehicle is in the preparatory depth determination stage, and the fuzzy PID proportional coefficient range combined from Table 2 and Table 3 is used to set the range;

(3) Maintain fixed depth: when Δy≤c, the current depth is approximately equal to the target depth, and no adjustment is required;

Among them: a, b, c are the allowable values of large depth error, small depth error and depth error, respectively, which are set according to the task requirements;

The setting range in the steps (1) and (2) depends on the correction of the PID controller parameters based on the fuzzy controller, that is, the fuzzy controller takes the depth deviation and the depth deviation rate of change as inputs, and uses the fuzzy controller to correct the PID control. device parameters;

The discrete universe of depth bias and depth bias change rate is {-3, -2, -1, 0, 1, 2, 3}, and the fuzzy language values are also {NB, NM, NS, ZE, PS, PM, PB}={negative big, negative middle, negative small, zero, positive small, positive middle, positive big};

Take Δy as the value of the abscissa E and _v as the value of the ordinate Ec to query the table, and the PID parameters after adjustment are:

Where: k _p is the original proportional coefficient, _ki is the original integral coefficient, k _d is the original differential coefficient; Δk _p is the proportional coefficient tuning adjustment amount obtained by the fuzzy controller looking up the table, Δk _i is the integral obtained by the fuzzy controller looking up the table Coefficient setting adjustment amount, Δk _d is the differential coefficient setting adjustment amount obtained by looking up the table of the fuzzy controller; k _pF is the proportional coefficient setting adjustment amount obtained by the fuzzy controller looking up the table. The ratio after adding the original proportional coefficient _{k p} coefficient, k _iF is the integral coefficient after adding the integral coefficient setting adjustment amount Δk _i obtained by the fuzzy controller look-up table and the original integral coefficient, k _dF is the differential coefficient setting adjustment amount obtained through the fuzzy controller look-up table Δk _d and The differential coefficient after the original differential coefficient Δk _d is added;

The depth control value Δy _c calculated by the PID controller with the PID parameters adjusted by the fuzzy controller is:

Among them, k _pF is the adjusted proportional coefficient, k _iF is the adjusted integral coefficient, k _dF is the adjusted differential coefficient, Δy is the deviation value from the current depth to the target depth, t is the integration time, and dt is the differential time;

Described table 1, table 2 and table 3 are:

Table 1: Large-scale correction table of _kp fuzzy control coefficient

Table 2: Small range correction table of _kp fuzzy control coefficient

Table 3: k _i /k _d fuzzy control coefficient correction table

Step 3: Discretize the depth control variable _Δyc to obtain the discrete control variable as:

Where: Δy _n is the depth deviation from the current depth to the target depth in the nth control cycle, and T is the discretization time interval;

Step 4. Calculate the actuator value according to the control amount:

Compute the position where the centroid block of the centroid mechanism moves:

M _c =K _M *Δy _c *(M _max -M _min )

Among them: K _M is the conversion coefficient from the control quantity to the execution value, M _max is the upper limit of the mass center block movement of the mass center mechanism, M _min is the lower limit of the mass center block movement of the mass center mechanism, Δy _c is the discrete control amount, and M _c is the mass center mechanism. The target position to be moved;

Calculate the magnitude of the caudal fin motion:

Q _c =R*(Q _max -Q _min )*sin(P*Δy _c )

Among them: P is the control variable conversion coefficient, R is the output gain coefficient, Q _max is the upper limit of the tail fin movement, Q _min is the lower limit of the tail fin movement, Δy _c is the discrete control variable, and Q _c is the target position of the tail fin movement;

Step 5: Control the mass center mechanism of the underwater vehicle with the target position M _c that the center of mass mechanism needs to move, and control the target position Q _c of the tail fin action to control the tail fin of the underwater vehicle, so as to change the pitch attitude angle of the aircraft under the flapping state of the pectoral fins, so that Pitching moments of different magnitudes are generated, so that the depth deviation of the underwater vehicle can reach the allowable error range, that is, the task of depth determination is completed during the fluttering process.

beneficial effect

The invention proposes a depth-fixing control method for a manta ray-like underwater vehicle based on the center of mass and caudal fin. For the pectoral fin propulsion bionic underwater vehicle with strong model uncertainty, the depth-fixing task is divided into three stages by segmentation. , combined with the fuzzy controller to implement online correction to the PID coefficients to realize the adjustment of the structure of the centroid and the tail fin, and then complete the task of swimming at a fixed depth. During the depth-fixing process, the pectoral fin mechanism of the underwater vehicle is driven by the steering gear and keeps flapping all the time. The steering gear outputs a constant amplitude and phase difference to generate thrust along the X-axis of the carrier coordinate system. The pitching moment is generated by changing the forward and backward movement of the mass center block and the up and down deflection of the tail fin to achieve the effect of fixing the depth.

Adopting the present invention has the following beneficial effects:

1. The traditional PID control algorithm has certain limitations. The parameters in the control algorithm are fixed values, which do not change with the system and do not have self-tuning characteristics. The invention divides the depth-fixing control of the aircraft into three working conditions according to the magnitude of the depth deviation, and designs different fuzzy control rules according to different working conditions to correct the PID parameters, which has parameter self-tuning properties, and the method can be generally applied to no The accurate dynamic model of robot control has strong applicability and portability.

2. Since the obtained control quantity is linearized for the center of mass mechanism, nonlinearly calculated for the tail fin, and then acts on the actuator, the combined action of the center of mass mechanism and the tail fin can reach the maximum mechanism when the depth deviation is large. The maximum output value achieved by the vertical speed of the vehicle can be obtained, the response is faster, and the speed of the depth error convergence is greatly accelerated; when the depth deviation is between the large depth deviation and the small depth deviation, only the tail fin is used for adjustment. , reducing the power consumption, making the vehicle have longer endurance, and thanks to the nonlinear calculation of the output of the tail fin, the control accuracy can be improved and the overshoot can be reduced.

Description of drawings

Fig. 1 is the principle diagram of the depth-fixing control method of pectoral fin-driven underwater vehicle of the present invention;

Fig. 2 is the flow chart of the depth determination procedure of the present invention.

Figure 3 is a schematic diagram of the imitation manta ray underwater vehicle. The flapping structures on both sides are symmetrical with respect to the main body of the vehicle. The meanings of the numbers in the figure are as follows:

1, 2, and 3 are the fin units of the aircraft flapping wing frame;

4 is the head of the aircraft;

5 is the main body of the aircraft, and the center of mass block, electronic warehouse, etc. are all wrapped in it;

6 is the flapping wing on the right side of the aircraft;

7 is the tail fin part of the aircraft.

Detailed ways

The present invention will now be further described in conjunction with the embodiments and accompanying drawings:

The technical scheme adopted by the present invention is to obtain the current depth information through the depth sensor, and to use the fuzzy PID control based on segmentation to adopt different control strategies in the three stages of rapid dive/ascent, starting the fixed depth and maintaining the fixed depth. The centroid structure and the tail fin are adjusted under the working conditions, and finally the depth control of the underwater vehicle in the flapping state is realized. The specific steps are as follows:

Step 1: Calculate the depth deviation and the rate of change of the depth deviation.

The current depth of the underwater vehicle obtained through the depth sensor is y (downward is positive), and the reference depth set by the task is y _d , then the depth deviation Δy is

Δy=yy _d (1)

Derivation of the depth deviation, the depth deviation change rate _v is obtained as

where t is the update period of the depth sensor information of the underwater vehicle.

Step 2: Execute the segmentation strategy.

Based on the complexity and parameter uncertainty of the pectoral fin propulsion bionic underwater vehicle model, an expert segmented controller is designed. After obtaining the real-time information of the depth deviation of the control system, the expert segmentation strategy is executed.

The segmented controller uses the depth deviation as input to divide the depth determination task into the following three conditions:

(1) Rapid dive/ascent: when Δy≥a, that is, the current depth is far from the target depth, the underwater vehicle is in the rapid dive/ascent stage, and the fuzzy control table is selected as Table 4 and Table 6. Under this condition, the center of mass mechanism and the tail fin act together as the actuator according to the control amount. At this time, the center of mass mechanism and the tail fin work together to generate a large pitching moment according to the control amount, so as to achieve rapid diving and ascending.

(2) Start depth determination: when a≥Δy≥b, the current depth is close to the target depth, the underwater vehicle is in the preparatory depth determination stage, and the fuzzy control table is selected as Table 5 and Table 6. In this case, only the tail fin part is adjusted according to the control amount. At this time, only the center of mass system returns to the neutral position, and only the tail fin is adjusted according to the control amount, resulting in a small pitching moment and reducing overshoot.

(3) Maintain fixed depth: when Δy≤c, the current depth is approximately equal to the target depth, and no adjustment is required.

Among them, a, b, and c are the set maximum depth error, small depth error and allowable value of depth error, respectively.

Step 3: Correct the PID controller parameters based on the fuzzy controller.

The fuzzy controller takes the depth deviation and the depth deviation rate as input at the same time, and uses the fuzzy controller to correct the parameters of the PID controller. The schematic diagram is shown in Figure 1.

The basic domain of depth bias is Δy∈[-|y _max |, |y _max |], where y _max is the maximum value of depth bias; the domain of variation rate of depth bias is v∈[-|v _max |, | v _max |], where v _max is the maximum value of the depth deviation rate. The discrete universe of depth bias and depth bias change rate is {-3, -2, -1, 0, 1, 2, 3}, and the fuzzy language values are also {NB, NM, NS, ZE, PS, PM, PB}={negative large, negative medium, negative small, zero, positive small, positive medium, positive large}.

According to different working conditions, the original PID parameters are self-tuned by using the fuzzy controller to perform the table lookup operation. The table query is carried out with Δy as the value of the abscissa E and _v as the value of the ordinate Ec. The adjusted PID parameters are:

In formula (3), k _p is the original proportional coefficient, _{k i} is the original integral coefficient, and _{k d} is the original differential coefficient; The integral coefficient tuning adjustment amount obtained from the table, Δk _d is the differential coefficient tuning adjustment amount obtained by looking up the table of the fuzzy controller; k _pF is the proportional coefficient tuning adjustment amount obtained by the fuzzy controller looking up the table Δk _p is in phase with the original proportional coefficient k _p The added proportional coefficient, k _iF is the integral coefficient after the addition of the integral coefficient tuning adjustment value Δki obtained by the fuzzy controller look-up table and the original integral coefficient, _{k dF} is the differential coefficient tuning adjustment obtained by the fuzzy controller look-up table The differential coefficient after the quantity Δk _d is added to the original differential coefficient Δk _d ;

The fuzzy control tables of the embodiment are shown in the attached tables 4, 5 and 6.

The depth control value Δy _c calculated by the PID controller with the PID parameters adjusted by the fuzzy controller is:

In formula (4), k _pF is the adjusted proportional coefficient, k _iF is the adjusted integral coefficient, k _dF is the adjusted differential coefficient, Δy is the deviation value from the current depth to the target depth, t is the integration time, dt is the differential time;

By discretizing the above formula, the discrete control quantity can be obtained as:

In formula (5), Δy _n is the depth deviation from the current depth to the target depth in the nth control cycle, and T is the discretization time interval.

Step 4: Calculate the actuator value according to the control amount.

According to the discretized control quantity, the moving position of the centroid block of the centroid mechanism can be calculated. The specific formula is:

M _c =K _M *Δy _c *(M _max -M _min ) (6)

In formula (6), K _M is the conversion coefficient from the control quantity to the execution value, M _max is the upper limit of the movement of the mass center block of the mass center mechanism, M _min is the lower limit of the mass center block movement of the mass center mechanism, Δy _c is the discrete control amount, M _c The target position that the centroid mechanism needs to move.

The amplitude of the tail fin action can be calculated according to the discretized control amount, and the specific formula is:

Q _c =R*(Q _max -Q _min )*sin(P*Δy _c ) (7)

In formula (7), P is the control variable conversion coefficient, R is the output gain coefficient, Q _max is the upper limit of the tail fin movement, Q _min is the lower limit of the tail fin movement, Δy _c is the discrete control variable, and Q _c is the target of the tail fin movement. Location.

For example, when M _max is 20, M _min is -20, M _c is 0, Q _max is 30, Q _min is 30, and Q _c is 5, it means that the execution value of the centroid mechanism is 0, and the centroid block should be kept at the zero position. Action; the execution value of the caudal fin is 5, and the caudal fin should be hit to the position of 5°, which is the start of the depth-fixing stage.

Adjust the calculated reference center of mass position of the underwater vehicle and the deflection angle of the caudal fin according to the control amount, adjust the center of mass mechanism and the caudal fin so that the center of mass and the caudal fin of the underwater vehicle reach the reference position, and change the pitch attitude angle of the aircraft when the pectoral fin flaps , so that the underwater vehicle generates a pitching moment, adjusts the depth, and the depth deviation reaches the allowable error range, that is, the depth setting process of the underwater vehicle in the state of pectoral fin flapping is realized. The program flow chart is shown in Figure 2. The schematic diagram of the manta ray underwater vehicle is shown in Figure 3.

Schedule 4:

Specific example k _p fuzzy control coefficient wide range correction table

Schedule 5:

Specific example k _p fuzzy control coefficient small range correction table

Schedule 6:

Specific example k _i /k _d fuzzy control coefficient correction table

Claims (1)

1. A method for controlling the depth of an artificial bat ray underwater vehicle based on a centroid and a tail fin is characterized by comprising the following steps:

step 1, calculating depth deviation and depth deviation change rate: acquiring the current depth of the underwater vehicle as y and the current depth is positive downwards through a depth sensor; the reference depth set by the task is y _d Then the depth deviation is Δ y:

Δy＝y-y _d

and (3) deriving the depth deviation to obtain a depth deviation change rate v as:

wherein: t is the depth sensor information updating period of the underwater vehicle;

step 2, taking the depth deviation as an input to divide the depth fixing task into the following three working conditions:

(1) rapid submergence/floatation: when delta y is larger than or equal to a, namely the current depth is far away from the target depth, the underwater vehicle is in a rapid submerging/floating stage, and a large fuzzy PID proportion coefficient setting range formed by combining the table 1 and the table 3 is adopted;

(2) starting and depth setting: when a is larger than or equal to delta y and larger than or equal to b, the current depth is close to the target depth, the underwater vehicle is in a prepared depth fixing stage, and a fuzzy PID proportional coefficient range setting range formed by combining a table 2 and a table 3 is adopted;

(3) keeping the depth to be fixed: when delta y is less than or equal to c, the current depth is approximately equal to the target depth without adjustment;

wherein: a, b and c are respectively depth large error, depth small error and depth error allowable values, and are set according to task requirements;

the setting range in the step (1) and the step (2) depends on the PID controller parameter corrected based on the fuzzy controller, namely the fuzzy controller simultaneously takes the depth deviation and the change rate of the depth deviation as input and corrects the PID controller parameter by using the fuzzy controller;

the discrete domain of depth deviation and the change rate of the depth deviation is { -3, -2, -1, 0, 1,2, 3}, and the fuzzy language value is { NB, NM, NS, ZE, PS, PM, PB }, i.e., { large negative, medium negative, small negative, zero, small positive, medium positive, large positive };

and performing table query by taking Δ y as the value of an abscissa E and v as the value of an ordinate Ec, wherein the corrected PID parameters are as follows:

wherein: k is a radical of _p Is the original scale factor, k _i For the original integral coefficient, k _d Is the original differential coefficient; Δ k _p The proportional coefficient setting adjustment, Δ k, is obtained by looking up a table for the fuzzy controller _i Integral coefficient setting adjustment, Δ k, obtained for fuzzy controller look-up table _d Obtaining a differential coefficient setting adjustment quantity for a fuzzy controller by looking up a table; k is a radical of _pF For the adjustment quantity delta k of the proportional coefficient setting obtained by the look-up table of the fuzzy controller _p Coefficient of proportionality to original k _p Summed scaling factor, k _iF Setting adjustment quantity delta k of integral coefficient obtained by looking up table through fuzzy controller _i And the original integral coefficient k _i Added integral coefficient, k _dF Setting adjustment quantity delta k of differential coefficient obtained by looking up table of fuzzy controller _d And the original differential coefficient k _d The added differential coefficients;

depth control quantity delta y calculated by PID controller for setting PID parameter by using fuzzy controller _c Comprises the following steps:

wherein k is _pF For adjusted proportionality coefficient, k _iF For adjusted integral coefficient, k _dF For the adjusted differential coefficient, Δ y is a deviation value from the current depth to the target depth, t is integration time, and dt is differentiation time;

the tables 1,2 and 3 are:

TABLE 1 k _p Large range correction table for fuzzy control coefficient

TABLE 2 k _p Fuzzy control coefficient small range correction table

TABLE 3 k _i /k _d Fuzzy control coefficient correction table

And step 3: for depth control quantity delta y _c The discretization treatment can obtain a discretized control quantity system as follows:

wherein: Δ y _n The depth deviation from the current depth to the target depth in the nth control period is shown, and T is a discretized time interval; k is the current nth cycle.

And 4, calculating an actuating mechanism value according to the control quantity:

calculating the position of the centroid block movement of the centroid mechanism:

M _c ＝K _M *Δy _c *(M _max -M _min )

wherein: k _M For conversion of control quantities into execution values, M _max Is the upper limit of the movement of the mass center mechanism, M _min Is the lower limit of the movement of the mass center mechanism, Δ y _c For discrete controlled variables, M _c The target position of the mass center mechanism needing to be moved;

calculating the amplitude of the tail fin action:

Q _c ＝R*(Q _max -Q _min )*sin(P*Δy _c )

wherein: p is the control quantity conversion coefficient, R is the output gain coefficient, Q _max Upper limit of motion of the tail fin, Q _min Lower limit of motion of the tail fin, Δ y _c For discrete controlled quantities, Q _c Is the target position of the tail fin action;

and 5: target position M of mass center mechanism needing to be moved _c Control of the centroid mechanism of an underwater vehicle, target position Q of the skeg motion _c The tail fin of the underwater vehicle is controlled, and the pitching attitude angle of the pectoral fin of the vehicle in a flapping state is changed, so that pitching moments with different sizes are generated, the depth deviation of the underwater vehicle reaches an error allowable range, and a depth fixing task is completed in the flapping process.

Images