CN107932489A - A kind of robot cycling device and control method - Google Patents

Abstract

The invention discloses a kind of robot cycling device and control method, comprise the following steps：Relation between car body roll angle and driving balance turbin generator output quantity is determined by Dynamic Modeling and analysis；Control car body to carry out curvilinear motion, observe the control response of car body balance when carrying out curvilinear motion under different driving speed, different handle angles, and therefore formulate controlled device target；Introductory path planning or tracking traveling scheme are formulated according to the situation of above-mentioned debugging；System is optimized and improved using intelligent control algorithm according to actual debugging situation.The compact simple and cost of bicycle machines people that the present invention is studied is relatively low, with stronger anti-interference, balance control for other related under-actuated systems such as the bicycle machines people of following other forms, wheelbarrow robot, scooter robot and lay certain basis.

Description

Device and control method for a robot riding a bicycle

technical field

The invention relates to a robot riding a bicycle device and a control method.

Background technique

In recent years, with the rapid development of the robot industry, especially based on the vigorous development of emerging sensors and control algorithms, various types of mobile robots can be well applied to various industrial production links and save a lot of manpower. Mobile robots with navigation systems can perform various surveillance, inspection and even simple tasks in some special working environments, such as underground, infectious disease hospitals, and nuclear facilities, where there are unsafe factors and humans are not allowed to enter directly. Interactive tasks. Most mobile robotic system chassis use four-wheel, multi-wheel or even tracked structures. The mobile robot system with four-wheel and multi-wheel structure will have static stability when it is stationary, but it has the disadvantages of requiring a large working space due to its large turning radius, inflexible action, complex structure, and high power consumption.

When encountering some narrow and long-distance workspaces, the multi-wheel robot will no longer be able to pass lightly and quickly. At this time, the mobile robot with two wheels longitudinally arranged in the traditional bicycle structure has once again attracted people's attention. Compared with ordinary four-wheel or multi-wheel mobile robots, although the bicycle robot system has underactuated instability (it needs to be assisted by the program to maintain dynamic balance), but once it is running, it is flexible and has a simple structure and a relatively small body. Narrow, so the weight is lighter and the energy consumption is small, which meets the long-term needs of energy conservation, emission reduction and energy consumption reduction in my country's current economic transformation period. After the control algorithm is mature, the bicycle robot system can also adapt to various complex road conditions, and has potential application prospects in security inspection, resource exploration and even logistics industry. At the same time, if the bicycle robot system is equipped with a humanoid robot from the frame to follow or drive the "pedals", it will also be entertaining and popular, and it will also have a certain potential market value.

At present, most of the bicycle robots made at home and abroad are mainly based on the bicycle body, that is, the target system is a robot-shaped bicycle, and the balance of the robot bicycle can be realized by designing a frame similar to a bicycle. The driving method is mainly driven by motor. During the operation of the bicycle, the system does not have other dynamic disturbances except for the adjustment of the handle of the bicycle body or the internal balance structure. The system designed in this application is to enhance the appreciation and promote the further development of humanoid robots. At the same time, in order to get closer to the concept of "bicycle robot", a "robot that can ride a bicycle", a humanoid robot will be used to actively drive the bicycle body Forward, its legs are designed with multi-joint steering gear, and the overall system is assisted by a mechanical auxiliary device installed on the car body to maintain balance. Bicycle robot is an incomplete and underactuated system with typical symmetry characteristics. The two wheels of the bicycle body are arranged longitudinally, and the lateral inclination angle is not directly driven. It is constrained by the second-order dynamic coupling. The nonlinear control of the structure to achieve the balance of the whole system is a recognized problem in the academic circles.

Contents of the invention

The purpose of the present invention is to provide a robot riding a bicycle device and a control method to solve the problems raised in the above-mentioned background technology.

To achieve the above object, the present invention provides the following technical solutions:

A method for controlling a robot riding a bicycle, comprising the following steps:

S1, determine the relationship between the roll angle of the car body and the output of the motor driving the balance wheel through dynamic modeling and analysis;

S2, control the car body to move in a curve, observe the control response of the car body balance when the car body moves in a curve at different driving speeds and different handle angles, and formulate the target of the controlled object accordingly;

S3, according to the debugging situation of S1 and S2, formulate relevant path planning or tracking driving plan;

S4, optimize and improve the system by using intelligent control algorithm according to the actual debugging situation.

Furthermore, in S1, the entire bicycle body is first simplified into a geometric model, the front handlebar is fixed to the neutral position and the rotation angle is 0 degrees, and the "robot" system is regarded as a fixed load fixed on the seat; through The position of the overall center of gravity of the car body is inferred from the linear superposition of the mass and geometric relationship of each part of the car body; according to the fixed position, moment of inertia and mass of the inertia wheel, it is deduced that when the inertia wheel rotates, the acceleration or reverse deceleration of the inertia wheel produces When the lateral force of the vehicle body rolls and dumps sideways, the moment equation of the component force of gravity acting on the vehicle body topple over laterally cancels out, so as to establish its dynamic model; according to the correlation function relationship, the traditional PID control algorithm is first used The controller is designed so that the controller can output the motor control amount in real time or in advance according to the roll angle of the vehicle body, the acceleration of the roll angle, or other detection quantities.

Further, in S2, the turning radius is deduced from the distance between the front and rear contact points of the wheel, the handlebar angle and other related quantities; the centripetal force of the system is calculated according to the driving speed, turning radius and the overall mass of the system when turning, and again according to The angle of inclination that should be maintained by the vehicle body can be obtained by equal torque; according to the relationship between the angle of roll angle that should be maintained by the vehicle body, the driving speed, and the handlebar angle, the traditional PID control algorithm is used to design the curve motion balance controller again, so that the controller can be based on The roll angle of the car body, the acceleration of the roll angle, the driving speed, the handlebar angle and other detected quantities output the motor control value in real time or in advance to keep the car body at a certain inclination angle and ensure that the curve movement can be realized smoothly and quickly.

Furthermore, in S3, if the car body is running relatively stable and the vibration and error are small when it is moving straight forward or turning, use the on-board high-precision encoder and gyroscope to calculate the current position in real time; if the actual debugging result has a large error Or to achieve tracking, use a linear array CCD to constantly correct the car body to drive along the color bar by detecting the color bar that is laid on the ground with a large difference in color from the background ground.

Furthermore, in S4, the intelligent control algorithm includes fuzzy parameter self-tuning PID control algorithm and parameter self-adaptive algorithm. At the same time, through communication with Bluetooth or WIFI module, use a computer or write an Android control APP to control through terminals such as mobile phones and tablets.

A robot bicycle riding device, using 32-way steering gear control board, through communication with STM32 single-chip microcomputer, STM32 single-chip microcomputer sends control commands, and each "leg" of the three steering gears "leg pedal" is divided into one cycle There are 20 position points, and the simultaneous control of multiple servos is realized by sending each position point in a periodic order.

Compared with the prior art, the beneficial effect of the present invention is: the application aims at the balance control problem of the bicycle robot, which is difficult to realize in a typical symmetrical underactuated non-holonomic constraint system, according to the mass of the vehicle body itself and its geometric relationship, through For the force decomposition and analysis of its dynamic model, the traditional PID controller is used for feedback control of the inertial wheel motor and the steering gear, and the infrared, ultrasonic or linear CCD sensor is used to avoid obstacles and travel along the predetermined trajectory. The actual situation of the debugging system uses advanced intelligent algorithm controllers such as fuzzy PID, and the system is connected to the wireless network through WIFI or Bluetooth module for easy control and use. At the same time, the multi-joint steering gear humanoid robot is used to realize the "leg kick" action to drive the bicycle, so that the bicycle robot can effectively realize the self-balancing of the system in ordinary straight-line driving. At the same time, controlling the direction of the bicycle robot is a steering gear mounted on the handlebar. When the bicycle moves straight forward at a relatively fast speed, if it is to turn quickly and realize the flexible curve movement of the vehicle body, it is necessary for the vehicle body to produce a certain inclination angle to offset the centripetal force during the curve movement. At this time, it is necessary to calculate the output value of the inertia wheel motor again according to the dynamic model, the forward running speed of the car body and the steering angle, so that the car body structure can maintain a certain inclination angle to perform a smooth and fast curve movement. Ultimately, real-time control of the system or driving along a predetermined trajectory can also be achieved.

In summary, the bicycle robot studied in this application is small, simple, low-cost, and has strong anti-interference performance. A certain basis.

Description of drawings

Fig. 1 is a structural schematic diagram of the present invention.

In the figure: 1-handlebar motor, 2-balance wheel, 3-control box with lithium battery, 4-steering gear.

Detailed ways

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

The research object of this application is the "robot cycling system", in which the periodic reciprocating action of "leg pedaling" to drive the bicycle is simply performed by the multi-joint steering gear robot. The communication of the board or use the MCU to directly output multiple PWM square wave control, and the pedal speed can be adjusted by adjusting the time required to execute a single action or the interval of multiple PWM square wave output signals, but if you want It is very difficult to achieve the balance control of the whole system in various motion situations such as static, straight line, curve, acceleration and deceleration, etc. At the same time, due to the simple structure of the bicycle frame, it is a slender front and rear, and a thin and narrow body in the lateral direction. There will be uncertain rolling and shaking, and the load capacity is limited, so the selection, installation and implementation of tracking or path planning sensors and data processors are also full of challenges.

Therefore, the main research goal of this application is to realize the balance control of the overall system under the dynamic disturbance of the periodic execution of the "humanoid robot leg pedal bicycle" action. Explore the relationship between the inclination angle of the car body and the angle of the steering handle when the car body is moving in a curve at a high speed, so as to realize the smooth and fast curve movement of the system. In addition, this application will also carry out path planning or tracking control on the "robot cycling" system, so that it can realize functions such as driving along a predetermined trajectory and returning to the starting point without human intervention.

The specific experimental method, research method and content are mainly divided into three steps.

Step 1: The main research method is to seek the relationship between the roll angle of the car body and the output of the motor driving the balance wheel through dynamic modeling and analysis. Firstly, the entire bicycle body is simplified into a geometric model, the front handlebar is fixed to the neutral position and the rotation angle is 0 degrees, and the "robot" system is regarded as a fixed load fixed on the seat. Through the linear superposition of the mass and geometric relationship of each part of the car body, the position of the overall center of gravity of the car body is deduced. According to the fixed position of the inertia wheel (the center of mass of the balance wheel), the moment of inertia and the mass, it is deduced that when the inertia wheel rotates, the lateral force generated by the acceleration or reverse deceleration of the inertia wheel is the same as when the vehicle body rolls sideways, and the gravity acts on the vehicle. The moment equation of the component forces of the lateral dumping on the body cancels out, so as to establish its dynamic model. According to the correlation function relationship, the traditional PID control algorithm is used to design the controller first, so that the controller can output the motor control value in real time or in advance according to the roll angle of the vehicle body, the acceleration of the roll angle, or other detection quantities. The lateral force generated by the rotation of the inertia wheel is offset by the dumping force, so that the posture correction of the lateral dumping of the vehicle body is realized.

The above research process can actually be simplified into a problem of a first-order inverted pendulum. Therefore, when the specific structure of the "robot cycling system" or the feasibility of its scheme has not been determined, a simulation can be designed in advance based on the estimated mass of each part corresponding to the "robot cycling" system and the geometric relationship determined by the design. First order inverted pendulum system.

Once the algorithm controlling the balance of the inverted pendulum stabilizes, the algorithm can be ported to the bicycle-riding robot. After the bicycle robot system can maintain balance whether it stops or moves forward and backward after starting, the design of the balance controller of the "bicycle robot" is completed. After completing the design of the balance controller of the system, the balance controller can still run stably under the dynamic disturbance of the "humanoid robot leg pedal drive" to keep the bicycle body within a certain roll angle of reciprocating motion. maintain balance.

Step 2: Try to make the car body move in a curve, observe the control response of the car body balance when the car body moves in a curve under different driving speeds and different handle angles, and formulate the target of the controlled object accordingly. Using the previously established dynamic model again, the front handlebar is fixed at a fixed angle, and the "robot" system is considered as a fixed load fixed on the saddle. Similarly, the turning radius is deduced from related quantities such as the distance between the front and rear contact points of the wheel, the handlebar angle, and the like. Calculate the centripetal force of the system according to the driving speed when turning, the turning radius and the overall mass of the system, and then obtain the angle of inclination that the car body should maintain according to the equal moment. According to the relationship between the angle of the car body and the driving speed and handlebar angle, the traditional PID control algorithm is used to design the curve motion balance controller again, so that the controller can control the roll angle of the car body, the acceleration of the roll angle, The driving speed, handlebar angle and other detected quantities output the motor control quantity in real time or in advance, so that the car body can maintain a certain inclination angle, and ensure that the curved movement can be realized smoothly and quickly.

Step 3: Make relevant route planning or tracking driving plan according to the debugging situation of the first two steps. If the car body is running relatively stable and the vibration and error are very small when it is moving forward in a straight line or turning, the current position can be calculated in real time using the on-board high-precision encoder and gyroscope. It is also conceivable to use a communication module and a photoelectric infrared sensor to cooperate with a robot cycling system to correct and input a fixed running track. If the error of the actual debugging result is large or tracking is to be realized, the linear array CCD can also be used to detect the color bar laid on the ground with a large difference in color from the background ground, and to always correct the car body to drive along the color bar. In addition to the linear array CCD that can be used for tracking, it is also possible to use the magnetic field navigation generated by burying the wires that are fed with alternating current.

Finally, according to the actual debugging situation, try to use intelligent control algorithms to optimize and improve the system, such as fuzzy parameter self-tuning PID control, parameter self-adaptive and other algorithms. At the same time, by communicating with the Bluetooth or WIFI module, using a computer or writing an Android control APP to control through mobile phones, tablets and other terminals, optimize the debugging process, and improve the control and user experience.

Since the "humanoid robot" can be regarded as a fixed load as a follower device, in order to reduce the difficulty of controlling the balance of the bicycle body through the rotation of the handlebar, the bicycle body system without the "humanoid robot" is used for debugging and system Realization, the initial design of the controller is based on the balance car with two wheels arranged left and right, using the existing sensor detection and Kalman filter algorithm, the two-wheel balance car is controlled by the multi-loop PID algorithm, but the two-wheel left The balance car arranged in the rear is not an underactuated system, and the balance control in the front and rear directions can be directly controlled by the motor-driven wheels, which is an ordinary first-order inverted pendulum control system. Therefore, the bicycle robot cannot directly obtain the relationship between the roll angle of the vehicle body and the handlebar angle using only one PID algorithm. Therefore, a multi-input and single-output simplified dynamics model is established using the Lagrangian method. According to the principle of partial feedback linearization, the underactuated subsystem including the vehicle body roll angle and handlebar torque is linearized, and A complex relationship was obtained, and finally the expected effect was tentatively achieved through simulation using the PID algorithm.

As shown in Figure 1, a motor and a balance wheel are installed on the back of the "humanoid robot", and the mechanical structure of the legs is changed to a multi-joint active drive structure realized by using multiple sets of steering gears. The handlebars are controlled by the rotation of the steering gears The direction can drive the upper limb mechanical structure of the "humanoid robot" to follow or deliberately embrace it to make a "big hand" shape. According to the designed system structure, the humanoid robot foot structure is assembled, and the multi-joint steering gear is debugged. Cooperating with the action algorithm of "leg pedaling the car", the scheme uses a 32-way steering gear control board. Through communication with the STM32 single-chip microcomputer, the STM32 single-chip microcomputer sends control commands, and each "leg" has three steering gear "leg pedals" The cycle of the pedal is divided into 20 position points, and the simultaneous control of multiple servos is realized by sending each position point in a periodic order, so that the ankle position at the end can drive the pedal to perform circular motion centered on the front axle of the front wheel disc .

It will be apparent to those skilled in the art that the invention is not limited to the details of the above-described exemplary embodiments, but that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Accordingly, the embodiments should be regarded in all points of view as exemplary and not restrictive, the scope of the invention being defined by the appended claims rather than the foregoing description, and it is therefore intended that the scope of the invention be defined by the appended claims rather than by the foregoing description. All changes within the meaning and range of equivalents of the elements are embraced in the present invention. Any reference sign in a claim should not be construed as limiting the claim concerned.

In addition, it should be understood that although this specification is described according to implementation modes, not each implementation mode only contains an independent technical solution, and this description in the specification is only for clarity, and those skilled in the art should take the specification as a whole , the technical solutions in the various embodiments can also be properly combined to form other implementations that can be understood by those skilled in the art.

Claims (6)

1. A method for controlling a robot riding a bicycle, comprising the following steps: S1, determine the relationship between the roll angle of the car body and the output of the motor driving the balance wheel through dynamic modeling and analysis; S2, control the car body to move in a curve, observe the control response of the car body balance when the car body moves in a curve at different driving speeds and different handle angles, and formulate the target of the controlled object accordingly; S3, according to the debugging situation of S1 and S2, formulate relevant path planning or tracking driving plan; S4, optimize and improve the system by using intelligent control algorithm according to the actual debugging situation. 2. The method for controlling a robot riding a bicycle according to claim 1 is characterized in that, in S1, at first the whole bicycle body is simplified into a geometric model, the front handlebar is fixed to the position of neutral position and rotation angle 0 degree, and " The "robot" system is regarded as a fixed load fixed on the vehicle seat; through the linear superposition of the mass and geometric relationship of each part of the vehicle body, the position of the overall center of gravity of the vehicle body is deduced; according to the fixed position, moment of inertia and mass of the inertia wheel, the When the inertial wheel rotates, the lateral force generated by the acceleration or reverse deceleration of the inertial wheel is counteracted by the component force of gravity acting on the vehicle body when the vehicle body rolls sideways, so as to establish its Dynamic model: According to the correlation function, the traditional PID control algorithm is used to design the controller first, so that the controller can output the motor control amount in real time or in advance according to the roll angle of the vehicle body, the acceleration of the roll angle, or other detection quantities. 3. The robot riding a bicycle control method according to claim 1 is characterized in that, in S2, the turning radius is deduced by related quantities such as the distance between the front and rear contact points of the wheel, the handlebar angle; according to the driving speed when turning , turning radius and the overall mass of the system to calculate the centripetal force of the system, and then obtain the inclination angle that the car body should maintain according to the equal moment; The traditional PID control algorithm designs the curve motion balance controller, so that the controller can output the motor control amount in real time or in advance according to the roll angle of the car body, the acceleration of the roll angle, the driving speed, the handlebar angle, and other detections, so that the car body maintains A certain inclination angle ensures that the curved movement can be realized smoothly and quickly. 4. The method for controlling a robot riding a bicycle according to claim 1, characterized in that, in S3, if the vehicle body runs relatively stably and the jitter and error are very small when moving forward in a normal straight line or turning, use a vehicle-mounted high-precision encoder and the gyroscope to calculate the current position in real time; if the actual debugging result has a large error or to achieve tracking, use a linear array CCD to correct the color of the car body at all times driving. 5. The robot bicycle riding device and control method according to claim 1, characterized in that, in S4, the intelligent control algorithm comprises a fuzzy parameter self-tuning PID control algorithm and a parameter self-adaptive algorithm, while communicating with Bluetooth or WIFI module , Use a computer or write an Android control APP to control through terminals such as mobile phones and tablets. 6. A robot cycling device, characterized in that it adopts a 32-way steering gear control board, communicates with the STM32 single-chip microcomputer, and the STM32 single-chip microcomputer sends control commands, and the 3 steering gears of each "leg" "leg kick" the foot The cycle of the pedal is divided into 20 position points, and the simultaneous control of multiple servos is realized by sending each position point in a periodic order.