Design and control method of wheel-leg balance robot capable of jumping actively

Abstract

The invention discloses a design and control method of a wheel-leg balance robot capable of actively jumping. The design of the wheel-legged balance robot is based on a planar parallel five-bar mechanism, the leg structure design and the wheel train structure design of the wheel-legged balance robot suitable for active jumping are comprehensively considered, and various sensor modules are adopted for sensing, so that the robot can autonomously jump on complex obstacle road sections; meanwhile, the diameter of the wheel and the torque of the motor are designed in a matched mode, so that the robot has high trafficability; a joint motor connecting structure is improved, so that the design strength of a leg structure is improved; an inverted pendulum model containing a wheel-leg mixed structure is used for fitting robot motion, an LQR control algorithm is adopted for processing a nonlinear system through piecewise linearization, and reference input is added to achieve trajectory tracking.

Description

Technical Field

[0001] The invention relates to the technical field of wheel-legged robots, and in particular to a design and control method of a wheel-legged balancing robot capable of active jumping. Background Art

[0002] The types and styles of mobile robots are becoming more and more mature, and they can be roughly divided into two categories: one is a foot-type walking robot and the other is a wheeled mobile robot. The foot-type walking robot has the advantages of anti-terrain interference and strong adaptability, but the complexity of its dynamic stability control limits the movement speed and practical application of the foot-type walking robot, and it cannot pass quickly when facing obstacles such as steps. Wheeled mobile robots have the advantages of fast movement speed and convenient load-bearing, but because of the limitations of the wheeled structure, they are more suitable for moving on flat ground and are helpless when facing complex gradient terrain structures. Therefore, on this basis, the foot-wheeled robot that combines the functions of the above two types of robots has attracted widespread attention. The use of the wheeled structure of the wheeled mobile robot combined with the foot-type mechanism of the foot-type walking robot can not only ensure that the robot has a high movement speed, but also enable the robot to move stably on complex terrain.

[0003] At present, many universities and scientific research institutions at home and abroad are conducting research on wheel-legged robots. After years of development, although the research on bipedal wheeled robots has made a series of progress, it is still a certain distance away from practical application. Bipedal wheeled robots are a kind of composite robot with multimodality. The main reason restricting the development of bipedal wheeled robots is that the performance of their system design and motion control cannot meet the needs of practical applications, especially the balance and stability control of the robot. Bipedal wheeled robots are naturally unstable systems. During work, they need to continuously exert control to maintain their own balance and stability. How the robot can maintain its own balance and stability while performing work tasks and resisting external interference are all key technologies. How to ensure rapid movement on complex terrain and gradient roads is the main problem studied in this invention.

[0004] The invention patent application with publication number CN 118579172 A discloses a symmetrical five-link closed-chain two-wheeled leg robot and its motion control method. The robot has a five-link closed-chain coupled drive wheel structure on one side, and adopts a left-right and front-back symmetrical body structure distribution. The robot described in the present invention adopts a six-degree-of-freedom structural design, with three degrees of freedom on each side, which respectively control the pitch of the body, the extension and retraction of the leg connecting rod and the rolling of the driving wheel. The dynamic balance control of the robot is completed through the coordinated cooperation of the hip joint motor and the wheel motor, and the rapid wheel movement on flat ground is realized. When encountering obstacles, the thigh connecting rod and the calf connecting rod are used to store energy through the leg retraction action and release energy through the extension action to achieve the jumping movement off the ground. However, this technical solution does not involve the design of the wheel system, the controller design and other design contents related to the rapid and stable movement of the robot on complex terrain and gradient roads. Summary of the invention

[0005] In view of the above-mentioned defects in the prior art, the present invention provides a design and control method for a wheel-legged balancing robot that can actively jump, so that the robot's leg structure is stronger, has stronger passability, and can autonomously jump over complex obstacle sections.

[0006] To achieve the above objectives, in a first aspect, the present invention provides a design of a wheel-legged balancing robot capable of active jumping, comprising a chassis, leg connecting rod structures on its left and right sides, a wheel set at the bottom of the leg connecting rod structure, power hardware and control hardware connected to the chassis;

[0007] The leg connecting rod structure is a planar parallel five-bar mechanism, including a frame fixed to the chassis, two thighs arranged outside the frame, and a front calf and a rear calf rotatably connected to the thighs; the stator of the hub motor in the wheel set is installed at the bottom of the front calf, and the corresponding position of the bottom of the rear calf is connected through a flange shaft to form a rotating pair; in order to allow the robot to be supported after power failure and fall to the ground, and the joint motor encoder can obtain the robot's tilting angle after power is turned on, a pair of guide wheels are respectively arranged on the front and rear sides of the chassis, and a laser ranging module is provided at the front of the bottom, and a positioning module is provided inside;

[0008] The thigh is connected to the rotor of the joint motor through the frame and the joint motor connector; the stator of the joint motor is fixedly connected to the chassis through a fixing plate; the thigh connected to the front calf is elevated by a raising plate to provide the vertical distance required for the calf connection to not cause interference; a leg limit block is provided at the connection between the thigh and the frame, wherein a limit raising plate is installed at the thigh connected to the front calf, which plays a role of compensation limit to ensure that the leg posture is determined after the robot is powered on again and does not lose the zero position.

[0009] Determine the wheel diameter range according to the initial speed and power requirements for hill climbing and the characteristic parameters of the hub motor to achieve matching;

[0010] The control hardware is two single-chip microcomputers, and two central boards are provided to distribute the power of the circuits.

[0011] The design of the wheel-legged balancing robot comprehensively considers the leg structure design and wheel system structure design of the wheel-legged balancing robot suitable for active jumping, and adopts a variety of sensor modules for perception, so that the robot can autonomously perform the jumping function on complex obstacle sections; at the same time, the wheel diameter and motor torque are matched in design, so that the robot has strong passability.

[0012] Furthermore, the main structure of the chassis includes a "well"-shaped frame composed of horizontal aluminum squares and vertical aluminum squares, and the robot lines are covered and shielded with plates; fixed aluminum squares and hardware mounting plates are fixedly connected to the "well"-shaped frame; the robot chassis is connected to the frame through a horizontal baffle to form an upper enclosure plate, and the two plates are connected by a four-hole adapter; the lower enclosure plate is composed of two identical lower side panels connected to the bottom side panel through a four-hole adapter to form a side enclosure; the robot is surrounded by four identical oblique anti-collision plates at the front and back; four identical bottom center side panels form the bottom enclosure, which are connected to the bottom center panel through a four-hole adapter to form a bottom enclosure structure.

[0013] Furthermore, the thigh and two calves are connected by plug bolts, and two pairs of flange bearings are embedded in the calves. The flange bearings bear radial force, and a thrust bearing is sandwiched between the pair of flange bearings to bear axial force; two gaskets are used to isolate the inner and outer rings of the flange bearings, and nuts are screwed on to form a rotating pair.

[0014] Furthermore, considering the moment of inertia, the length ratio of the thigh and calf is designed to be suitable for jumping, and the topology of the plate is optimized to reduce the mass and stress concentration.

[0015] Furthermore, in order to meet the requirement of the robot climbing over a slope, the peak torque T of the hub motor is max The following relationship is satisfied between the wheel diameter of the wheel set: , where θ is the flight slope, m is the mass of the robot, and T max is the peak torque, R max is the maximum diameter of the wheel set, g is the gravity constant; there is the following relationship between the initial speed v and the wheel diameter: , where R min is the minimum diameter of the wheel set, n is the wheel speed; there is the following relationship between the slope power P and the motor torque T: ; Given θ, m, T max, v, P, and n, the diameter range of the wheelset can be solved.

[0016] Furthermore, the wheel set includes a two-layer wheel rim and a hub motor, the outer ring of which is covered with rubber to form a rubber ring and is clamped by an inner plate and an outer plate; the inner ring is filled with a support member; the inner plate and the outer plate are connected with bolts through uniformly distributed through holes on the wheel rim; the hub motor is connected to the outer plate through a pin coupling and a load-reducing plate with a diameter larger than the diameter of the motor bolt distribution circle to form a motor rotor; the motor stator is connected to the front calf through a motor connector.

[0017] Furthermore, the guide wheel is clamped between the guide wheel outer mounting plate and the guide wheel inner mounting plate, the middle gap part of the guide wheel is padded with a pad, and the guide wheel is fixed by mounting bolts and mounting nuts; the guide wheel inner mounting plate is inserted into the guide wheel upper support plate and the guide wheel lower support plate through a plate insertion structure, and the guide wheel outer mounting plate is connected to the guide wheel upper support plate and the guide wheel lower support plate through a four-hole adapter.

[0018] Furthermore, the joint motor connecting part includes an inner ring plate with a plurality of first through holes on the ring surface and an outer ring plate with a plurality of second through holes on the ring surface; the joint motor connecting part is externally connected to a cross roller bearing and a bearing housing; the first through holes are used to connect the motor rotor and the leg, corresponding to the pin holes and bolt holes of the joint motor respectively; the second through holes are used to connect the leg connecting rod.

[0019] With this arrangement, the torque generated by the motor and the moment it is subjected to are borne by the pin, while the bolt only serves as a connection and is not subject to shear force, so it is not easy to break; the force exerted on the leg is borne by the bearing instead of being transmitted to the motor, thus avoiding wear of the thread and ensuring structural strength.

[0020] In a second aspect, the present invention further provides a control method for a wheel-legged balancing robot capable of active jumping, which adopts the above-mentioned wheel-legged balancing robot capable of active jumping, and comprises the following processes:

[0021] The robot system model is simplified to an inverted pendulum model, and the change in the robot leg length is ignored. The defined state vector includes variables related to the posture of the robot's upper structure and legs and the movement of the driving wheels; the variables include but are not limited to the body posture angle, the pendulum angle, and the angle of the line connecting the driving wheel shaft and the center of the motor shaft of the two joints of the leg relative to the inertial system;

[0022] According to the above wheel-leg inverted pendulum model, the control law is designed as a linear combination of the system states, namely:

[0023] ,in, is the state vector, is the control vector; is the feedback gain;

[0024] The feedback gain is calculated using a linear quadratic regulator (LQR) and the reference input is added to the input of the linear quadratic regulator, that is:

[0025]

[0026] The reference input The robot position is expected constitute:

[0027]

[0028] The linear quadratic regulator is linearized every 10 mm within the leg length range, and its feedback gain matrix is ​​solved; a polynomial equation is fitted to each element of the matrix as the leg length changes, and the following is obtained:

[0029] ;

[0030] In the formula, is the gain matrix related to leg length, When L0=0, the matrix K ij The value of , which is the constant term of the polynomial, is the coefficient of the first-order derivative of the matrix elements as the leg length L0 changes, is the coefficient of the second-order derivative of the matrix elements as the leg length L0 changes, is the feedback gain matrix K ij The coefficient of the third-order derivative of the elements of with the leg length L0, For leg length;

[0031] It is deduced that the robot longitudinal motion control law is:

[0032] ;

[0033] When the robot enters the active jumping mode, the laser ranging module 54 senses the obstacle in front and automatically reproduces the jumping control code. The four joint motors work together to control the leg retraction-extension-retraction to complete the jumping action.

[0034] Furthermore, the inverted pendulum model adopts a segmented modeling approach, firstly establishing dynamic equations for the driving wheel, the pendulum rod and the body respectively, and then combining these equations to obtain a state space model by eliminating intermediate variables.

[0035] Compared with the prior art, the above invention has the following advantages or beneficial effects:

[0036] (1) The present invention takes into account the characteristics of both wheeled and legged robots, and designs a wheel-legged balancing robot based on a planar parallel five-bar mechanism; the joint motor connection structure is improved to increase the design strength of the leg structure. By calculating the matching relationship between the wheel diameter and the motor torque of the wheel-legged robot, they are better matched, so that the robot has stronger passability.

[0037] (2) The present invention innovatively uses an inverted pendulum model containing a wheel-leg hybrid structure to fit the robot motion, and adopts the LQR control algorithm to process the feedback gain matrix under different leg lengths by polynomial fitting, piecewise linearization to process the nonlinear system, and adds reference input to achieve trajectory tracking, which makes progress in the related motion control methods in the field of bipedal wheeled robot motion control, such as balance, obstacle crossing, multimodal movement, center of gravity adaptation, etc.

[0038] (3) The present invention combines multiple sensor modules with innovative control methods to effectively complete the control of autonomous jumping. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The present invention and its features, appearance and advantages will become more apparent by reading the detailed description of non-limiting embodiments with reference to the following drawings. The same reference numerals indicate the same parts in all the drawings. The drawings are not drawn completely to scale, and emphasis is placed on illustrating the subject matter of the present invention.

[0040] Figure 1 A schematic diagram of the overall structure of a wheel-leg balancing robot provided by an embodiment of the present invention;

[0041] Figure 2 A schematic diagram of a top view of the wheel-leg balancing robot provided in an embodiment of the present invention;

[0042] Figure 3 A schematic diagram of the outer side structure of one leg of a robot provided by an embodiment of the present invention;

[0043] Figure 4 A schematic diagram of the inner side structure of one leg of a robot provided by an embodiment of the present invention;

[0044] Figure 5 A schematic diagram of the connection structure of the connection between the thigh and calf of a robot provided by an embodiment of the present invention;

[0045] Figure 6 A schematic diagram of the structure of a robot thigh provided by an embodiment of the present invention;

[0046] Figure 7 A schematic diagram of the structure of the robot front calf provided by an embodiment of the present invention;

[0047] Figure 8A schematic diagram of the structure of the robot hind leg provided by an embodiment of the present invention;

[0048] Fig. 9 A schematic diagram of the structure of a robot joint motor connecting member provided by an embodiment of the present invention;

[0049] Fig.10 A schematic cross-sectional view of a robot joint motor connection structure provided by an embodiment of the present invention;

[0050] Fig.11 A schematic diagram of the structure of a robot wheel assembly provided by an embodiment of the present invention;

[0051] Fig.12 A schematic diagram of a central cross-sectional structure of a robot wheel assembly provided by an embodiment of the present invention;

[0052] Fig.13 A schematic diagram of a simplified model of a robot motion system provided by an embodiment of the present invention;

[0053] Among them, 1. wheel set; 2. front calf; 3. rear calf; 4. plug bolt; 5. gasket; 6. thigh; 7. frame; 8. leg limit block; 9. limit pad plate; 10. pad plate; 11. joint motor; 12. fixing plate; 13. fixing aluminum square; 14. flange shaft; 15. thrust bearing; 16. flange bearing; 17. nut; 18. horizontal aluminum square; 19. longitudinal aluminum square; 20. outer plate; 21. load reduction plate; 22. pin coupling; 23. drive wheel motor; 24. Motor connector; 25, wheel rim; 26, support; 27, rubber ring; 28, inner plate; 31, transverse baffle; 32, guide wheel upper support plate; 33, guide wheel lower support plate; 34, guide wheel outer mounting plate; 35, guide wheel inner mounting plate; 36, guide wheel; 37, pad column; 38, mounting bolt; 39, mounting nut; 40, reinforced aluminum square; 41, DC power supply; 42, power supply support frame; 43, four-hole adapter; 44, oblique anti-collision plate; 45, center plate; 46, STM32F4 Single chip microcomputer; 47, hardware installation plate; 48, bottom side plate; 49, bottom center plate; 50, cushion block; 51, anti-collision plate; 52, bottom center side plate; 53, laser ranging module mounting bracket; 54, laser ranging module; 55, RFID module; 56, lower side plate; 57, cover plate; 58, positioning module; 59, inner ring plate; 591, first through hole; 592, second through hole; 60, outer ring plate; 61, cross roller bearing; 62, bearing housing. DETAILED DESCRIPTION

[0054] The structure of the present invention is further described below in conjunction with the accompanying drawings and specific embodiments, but is not intended to limit the present invention.

[0055] In the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper", "lower", "front", "back", "left", "right", etc. are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, features defined as "first", "second", etc. may explicitly or implicitly include one or more of the features.

[0056] In the description of this application, it is also necessary to explain that, unless otherwise clearly specified and limited, the terms "setting" and "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; the method can be welding, threaded connection or binding connection; it can be directly connected or indirectly connected through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances. The standard parts used can be purchased from the market, and can be customized according to the description and drawings. The specific connection methods of each part adopt conventional means such as mature bolts, rivets, welding, etc. in the prior art. Machinery, parts and equipment all adopt conventional models in the prior art, and the structures and principles of the components known to technical personnel in this field can be known by technical personnel through technical manuals or through conventional experimental methods.

[0057] In order to make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0058] Example

[0059] See also Figure 1 and 2 , This embodiment provides a design of a wheel-leg balancing robot capable of active jumping, comprising a chassis, leg connecting rod structures on its left and right sides, a wheel set 1 at the bottom of the leg connecting rod structure, and power hardware and control hardware connected to the chassis;

[0060] The leg connecting rod structure is a planar parallel five-bar mechanism, including a frame 7 fixed to the chassis, two thighs 6 arranged on the outside of the frame 7, and a front calf 2 and a rear calf 3 rotatably connected to the thighs 6; the stator of the hub motor in the wheel set 1 is installed at the bottom of the front calf 2, and the corresponding position of the bottom of the rear calf 3 is connected through a flange shaft 14 to form a rotating pair; in order to allow the robot to be supported after power failure and fall to the ground, and the joint motor encoder can obtain the robot's tilting angle after power is turned on, a pair of guide wheels 36 are respectively arranged on the front and rear sides of the chassis, and a laser ranging module 54 is provided at the front of the bottom, and a positioning module 58 is provided inside;

[0061] The thigh 6 is connected to the rotor of the joint motor 11 through the frame 7 and the joint motor connector; the stator of the joint motor 11 is fixedly connected to the chassis through the fixing plate 12; the thigh 6 connected to the front calf 2 is raised by the raising plate 10 to provide the vertical distance required for the calf connection to not cause interference;

[0062] Determine the wheel diameter range according to the initial speed and power requirements for hill climbing and the characteristic parameters of the hub motor to achieve matching;

[0063] The control hardware is two single-chip microcomputers 46, and two central boards 45 are provided to distribute the power of the circuits.

[0064] More specifically, in this embodiment, see Figures 3 to 5 The leg connecting rod structure of the wheel-legged robot is a planar five-degree-of-freedom parallel five-bar mechanism. The connecting rods of the upper and lower arms on the left and right sides are centrally symmetrical, and the rod length ratios are the same.

[0065] The leg mechanism is connected to the "well"-shaped frame composed of the horizontal aluminum square 18 and the vertical aluminum square 19 through the fixed aluminum square 13, forming the main structure of the chassis. As the whole of the robot chassis, a plate is also used to cover and shield the robot line. The robot chassis is connected to the frame 7 through the horizontal baffle 31 to form an upper enclosure plate, and the connection between the two plates is connected by a four-hole adapter 43. The lower enclosure plate is composed of two identical lower side panels 56 connected to the bottom side panel 48 through a four-hole adapter 43 to form a side enclosure. The robot is surrounded by four identical oblique anti-collision plates 44 at the front and back; four identical bottom center side panels 52 form the bottom enclosure, which is connected to the bottom center panel 49 through a four-hole adapter 43 to form a bottom enclosure structure.

[0066] The planar parallel five-bar mechanism consists of a thigh 6, a front calf 2, a rear calf 3, and a frame 7The joint motor frame is composed of the following components: from bottom to top, the front shank 2 is used as the mounting plate of the hub motor stator in the wheel set 1, and the rear shank 3 is connected through the flange shaft 14 to form a revolute pair. The thigh 6 is connected to the two shanks by plugging bolts 4. Two pairs of flange bearings 16 are embedded in the shanks. The flange bearings bear radial forces. A thrust bearing 15 is sandwiched between the two pairs of flange bearings 16 to bear axial forces. Two gaskets 5 are used to isolate the inner and outer rings of the flange bearings 16, and nuts 17 are screwed on to form a revolute pair. The connection between the front shank 2, the rear shank 3 and the wheel set 1 is similar to the connection between the thigh and the leg, wherein the end face of the flange shaft 14 is threaded with a threaded hole for bolt tightening and axial positioning. The two thighs 6 are exactly the same. Since the shank connection requires a certain vertical distance, the left thigh is raised by the padding plate 10. At the same time, in order to ensure that the leg limit block 8 can limit the left thigh, a limit padding plate 9 is set, and the installation direction is parallel to the left thigh. During the rotation of the front calf 2, the limit pad plate 9 rotates synchronously to the leg limit block 8, playing the role of compensating the limit, so as to ensure that the leg posture is determined after the robot is powered on again, and the zero position is not lost. A countersunk hole is provided at the upper end of the thigh 6, which is used to connect the countersunk bolt to the joint motor connector and is connected to the rotor of the joint motor 11. The stator of the joint motor 11 is connected and fixed to the fixed aluminum square 13 through the fixing plate 12. The fixed aluminum square 13 is fixedly connected to the frame 7. Preferably, the front calf 2, the rear calf 3, the thigh 6, the frame 7, the limit pad plate 9, the pad plate 10, and the fixed plate 12 are all made of carbon fiber materials, and the leg limit 8 is made of 6061 aluminum alloy.

[0067] This embodiment also provides a relatively detailed design of the structure of the leg connecting rod.

[0068] According to the method of determining proportion, simulation and size, a more reasonable pole length ratio suitable for jumping is obtained. According to the rigid body moment of inertia calculation formula,

[0069]

[0070] In the formula - moment of inertia of the leg link;

[0071] ——connecting rod quality;

[0072] ——The vertical distance from the particle to the center of rotation.

[0073] It can be seen that the moment of inertia is not only related to the mass size, but also to the mass distribution. At the same time, according to the parallel axis theorem:

[0074]

[0075] In the formula ——Moment of inertia of the connecting rod’s center of mass around its center of mass; ——connecting rod quality;

[0076] ——The vertical distance between the connecting rod center of mass and the motor shaft.

[0077] It can be concluded that when the center of mass of the connecting rod is closer to its rotation axis, the moment of inertia of the connecting rod is smaller, and the torque required by the motor to overcome the mass of the connecting rod is smaller. Based on the above analysis, the thigh 6 connecting rod structure is designed as follows Figure 6 As shown, the lower legs 2 and 3 are designed as Figure 7 and 8 , and perform topological optimization of the plate to reduce mass and stress concentration.

[0078] See also Fig. 9 and Fig.10 The joint motor connecting part uses an integrated structural part open sourced by Shanghai Jiao Tong University, including an inner ring plate 59 with a plurality of first through holes 591 on the ring surface and an outer ring plate 60 with a plurality of second through holes 592 on the ring surface; the joint motor connecting part is externally connected to a cross roller bearing 61 and a bearing housing 62; the first through hole 591 is used to connect the motor rotor and the leg, corresponding to the pin hole and bolt hole of the joint motor respectively; the second through hole 601 is used to connect the leg connecting rod.

[0079] More specifically, the six circular holes on the inner circle of the joint motor connecting part are used to connect the motor rotor and the legs. The six holes correspond to the three pin holes and three bolt holes of the joint motor respectively, so that the torque generated by the motor and the moment it is subjected to are borne by the pins, while the bolts only play a connecting role, so that the bolts are not subjected to shear force and are not easy to break; the six threaded holes on the outer circle of the structural part are used to connect the leg connecting rods, and at the same time, the flange edge boss on the outermost circle also plays the role of positioning the inner ring of the cross-stick bearing.

[0080] A cross roller bearing 61 is added to the connection between the motor and the leg, so that the force on the leg is borne by the bearing instead of being transmitted to the motor, thus avoiding wear of the thread and ensuring the structural strength. In the specific bearing installation, the inner and outer rings are positioned separately, the inner ring is acted upon by the flange edge of the motor connector, and the outer ring is limited by the bearing housing.

[0081] See also Fig.11 and Fig.12In the wheel set 1, the outer ring of the wheel rim 25 is covered with a polyurethane material with a Shore hardness of 55A, and a rubber ring 27 is formed after the covering. The material of the wheel rim 25 is 6061 aluminum alloy, and it is clamped with the outer plate 20 through the inner plate 28. The inner ring of the wheel rim 25 is filled with a support member 26, so that the inner plate 28 and the outer plate 20 are connected with bolts through the uniformly distributed through holes on the wheel rim 25. The hub motor 23 is connected to the outer plate 20 through a pin coupling 22 and a load-reducing plate 21 with a larger diameter than the motor bolt distribution circle diameter, and the motor rotor connection is completed. The motor stator is connected to the left thigh 2 through the motor connector 24, and the stator is fixed.

[0082] The design of the wheel system is particularly important in the entire robot design. The size of the drive wheel directly affects the establishment of the robot's dynamic model. The selection of the hub motor and the wheel diameter are particularly critical for matching the torque and speed of the robot's wheel system. If a hub motor with a larger torque is selected, its speed will be small, and it may not be possible to reach the required speed when jumping over obstacles. Therefore, it is necessary to increase the wheel radius. However, increasing the wheel radius will increase the torque that the robot needs to overcome its own weight when going uphill. The formula is as follows:

[0083]

[0084] θ is the flying slope, m is the mass of the robot, T max is the peak torque, R max is the maximum diameter of the wheel set, and g is the gravity constant. If the flying slope θ is 17°, T max is the peak torque. It can be seen that if the wheel diameter is increased, the required torque will also increase, and if the weight is reduced, the required torque will decrease. At the same time, it is also necessary to meet the initial speed of the flying slope greater than 2m / s, and leave a certain safety factor. The design parameter is set to 2.5m / s, based on the power:

[0085] Required speed:

[0086] Where T is the torque, R min is the minimum diameter of the wheel set, and n is the wheel speed.

[0087] Combined with the characteristic parameters of the motor, the optimal range of wheel diameter can be determined to achieve a match.

[0088] As an example, this embodiment comprehensively considers the effects of being too large or too small, and the outer diameter of the driving wheel set of the robot is 220 mm.

[0089] Preferably, an RFID module 55 is also installed on the bottom center plate, which can be used to identify the positioning tag during the jump. Of course, for the use scenario without the positioning tag, the robot may not include this module.

[0090] As the hardware part of the robot, its power source is a 24V DC power supply 41. To prevent the power supply from being hit by uneven objects on the ground, an anti-collision plate 51 is set under the battery, connected to the guide wheel support lower plate 33 through a pad 50, and the battery is installed on the longitudinal aluminum square 19 through a battery support frame 42. The robot's controller is two STM32F4 microcontrollers 46, and two center boards 45 are set to distribute the power of the circuit. The robot hardware module is installed on the hardware installation board 47.

[0091] In order to allow the robot to have support after it loses power and falls to the ground, and to allow the joint motor encoder to obtain the robot's tilting angle after power is restored, two pairs of front and rear guide wheels 36 are set. The guide wheel 36 is clamped between the guide wheel outer mounting plate 34 and the guide wheel inner mounting plate 35. The middle gap of the guide wheel is padded with a cushion column 37, and the guide wheel is fixed by mounting bolts 38 and mounting nuts 39. The guide wheel inner mounting plate 35 is inserted into the guide wheel upper support plate 32 and the guide wheel lower support plate 33 through a plate plug structure, and the guide wheel outer mounting plate is connected to the guide wheel upper support plate 32 and the guide wheel lower support plate 33 through a four-hole adapter 43. In order to improve the structural rigidity of the aluminum square frame, a reinforced aluminum square 40 is set on the guide wheel upper mounting plate. Two pairs of guide wheels are set on the front and rear of the chassis, and the structure is exactly the same.

[0092] The control method of the wheel-legged balancing robot includes the following steps:

[0093] The robot system model is simplified to an inverted pendulum model, and the change in the robot leg length is ignored. The defined state vector includes variables related to the posture of the robot's upper structure and legs and the movement of the driving wheels; the variables include but are not limited to the body posture angle, the pendulum angle, and the angle of the line connecting the driving wheel shaft and the center of the motor shaft of the two joints of the leg relative to the inertial system;

[0094] According to the above wheel-leg inverted pendulum model, the control law is designed as a linear combination of the system states, namely:

[0095] ,in, is the state vector, is the control vector; is the feedback gain;

[0096] The feedback gain is calculated using a linear quadratic regulator (LQR) and the reference input is added to the input of the linear quadratic regulator, that is:

[0097]

[0098] The reference input The robot position is expected constitute:

[0099]

[0100] The linear quadratic regulator is linearized every 10 mm within the leg length range, and its feedback gain matrix is ​​solved; a polynomial equation is fitted to each element of the matrix as the leg length changes, and the following is obtained:

[0101] ;

[0102] In the formula, is the gain matrix related to leg length, When L0=0, the matrix K ij The value of , which is the constant term of the polynomial, is the coefficient of the first-order derivative of the matrix elements as the leg length L0 changes, is the coefficient of the second-order derivative of the matrix elements as the leg length L0 changes, is the feedback gain matrix K ij The coefficient of the third-order derivative of the elements of with the leg length L0, For leg length;

[0103] It is deduced that the robot longitudinal motion control law is:

[0104] ;

[0105] When the robot enters the active jumping mode, the laser ranging module 54 senses the obstacle in front and meets the parameter requirements in the wheel-legged robot jumping control model. The jumping control code is automatically reproduced, and the four joint motors 11 work together to control the leg retraction-extension-retraction to complete the jumping action.

[0106] More specifically, see Fig.13 Due to the complexity of the planar five-link model in dynamics and kinematics analysis, the robot motion system model is simplified to an inverted pendulum model as shown in the figure below. This model mainly focuses on the posture of the robot's upper mechanism and legs and the movement of the driving wheels, and ignores the change in the robot's leg length. It only considers the posture of the legs, that is, the angle between the driving wheel axis and the center of the motor shaft of the two joints of the leg relative to the inertial system.

[0107] Table 1 Definition of variables and parameters in the simplified model of the robot motion system

[0108]

[0109] The traditional inverted pendulum model usually only considers a single pendulum or double pendulum structure, while the model of this solution is a wheel-leg hybrid structure, including a driving wheel, a swing rod (the connection between the driving wheel shaft and the leg mechanism shaft) and a body (upper mechanism), which makes the system have more degrees of freedom and more complex dynamic equations. See Table 1. Different from the traditional inverted pendulum directly establishing the overall Lagrangian equation, this embodiment adopts a segmented modeling method:

[0110] Construct the dynamic equation for the driving wheel:

[0111]

[0112]

[0113]

[0114] Formulate the dynamic equations for the pendulum:

[0115]

[0116]

[0117]

[0118] Construct the dynamic equation for the body:

[0119]

[0120] State-space model:

[0121] Define the state vector With control vector They are:

[0122]

[0123] Define the nonlinear model of the system:

[0124]

[0125] Using MATLAB symbolic operation tools, according to equations (1.4), (1.5), (1.7), and (1.8), eliminate the intermediate variables. , and use the solve function to solve equations (1.3), (1.6), and (1.9) to obtain the symbolic expression of the system nonlinear model. According to the state vector With control vector , find the Jacobian matrix at the equilibrium point of the nonlinear model to linearize it, that is:

[0126]

[0127] in is the equilibrium point of the system, that is, the equation Solution:

[0128]

[0129] have:

[0130]

[0131] Since the expression is relatively complex and takes up a lot of space, symbols are used here instead.

[0132] All state variables can be directly measured or fused to obtain analytical results, so the system output for:

[0133]

[0134] in is a 6-dimensional unit matrix.

[0135] By substituting the model parameters into the state space model, the state matrix is ​​determined. and control matrix , its controllable matrix is ​​full rank, and the system is controllable. The system output matrix For the unit matrix, the system is obviously impressive.

[0136] According to the above wheel-leg inverted pendulum model, the control law is designed as a linear combination of the system states, namely:

[0137]

[0138] The linear quadratic regulator (LQR) is used to calculate the feedback matrix and define the cost function for:

[0139]

[0140] To make the cost function Minimum, input Should meet:

[0141]

[0142] Feedback Gain satisfy:

[0143]

[0144] in Satisfies the algebraic Riccati equation:

[0145]

[0146] The above method can achieve system stability near the linearized equilibrium point. In order for the robot to track the trajectory, a reference input must be added to the system input, namely:

[0147]

[0148] The reference input The robot position is expected constitute:

[0149]

[0150] In order to consider the working conditions of robots with different leg lengths, the system model is linearized every 10 mm in the leg length range, and its feedback gain matrix is ​​solved. That is, the nonlinear system is processed by piecewise linearization, and the feedback gain matrix K of each linearization point is solved separately. The polynomial equation is fitted to each element of the matrix as the leg length changes, and the following is obtained:

[0151]

[0152] The formula is derived, and the robot longitudinal motion control law is:

[0153]

[0154] The above control method enables the controller to better adapt to the working state of the wheel-legged robot with different leg lengths, and its implementation can be verified by building a simple simulation model through matlab simscape multibody.

[0155] Substituting the simulation model structural parameters, we can get System model in state:

[0156]

[0157] Select the LQR weight matrix as:

[0158]

[0159] Solving the Riccati equation gives the gain matrix:

[0160]

[0161] Let the robot track the step speed expectation :

[0162]

[0163] Integrating it can get the expected position of the robot , then the robot's expected state vector for:

[0164]

[0165] Take a field with positioning marks as an example. When the robot is moving, the RFID module 55 senses the tag on the ground, and the robot enters the automatic jumping mode. When the laser ranging module 54 fixed on the laser ranging module mounting frame 53 senses the obstacle in front, the jumping control code is automatically reproduced, and the four joint motors 11 work together to control the leg retraction-leg extension-leg retraction to complete the jumping action. After the jump is completed, the RFID module 55 detects the tag of the jump completion, unlocks the manual mode, and switches the control mode. At this time, the robot can be manually operated by the remote control. When the robot is active in the field, it can be positioned by the UWB positioning module 59 on the cover plate 58, and the position of the robot during movement can be known in the upper computer.

[0166] It should be noted that the device structure and drawings of the present invention mainly describe the principle of the present invention. In terms of the technology of the design principle, the settings of the power mechanism, power supply system and control system of the device are not fully described. On the premise that the technical personnel in this field understand the principle of the above invention, the details of its power mechanism, power supply system and control system can be clearly known. The control method of the application document is automatic control through a controller, and the control circuit of the controller can be realized by programming by technical personnel in this field.

[0167] Those skilled in the art should understand that those skilled in the art can implement variations by combining the prior art and the above embodiments, which will not be described in detail here. Such variations do not affect the essential content of the present invention, and will not be described in detail here.

[0168] The above describes the preferred embodiments of the present invention. It should be understood that the present invention is not limited to the above-mentioned specific embodiments, and the devices and structures that are not described in detail should be understood to be implemented in a common manner in the art; any technician familiar with the art can use the above-disclosed methods and technical contents to make many possible changes and modifications to the technical solutions of the present invention without departing from the scope of the technical solutions of the present invention, or modify them into equivalent embodiments of equivalent changes, which does not affect the essential content of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solutions of the present invention are still within the scope of protection of the technical solutions of the present invention.

Claims

1. A design of a wheel-leg balancing robot capable of active jumping, comprising a chassis, leg connecting rod structures on the left and right sides thereof, a wheel set (1) at the bottom of the leg connecting rod structure, power hardware and control hardware connected to the chassis; characterized in that: The leg connecting rod structure is a planar parallel five-bar mechanism, comprising a frame (7) fixed to the chassis, two thighs (6) arranged outside the frame (7), and a front calf (2) and a rear calf (3) rotatably connected to the thighs (6); the stator of the hub motor in the wheel set (1) is installed at the bottom of the front calf (2), and the corresponding position of the bottom of the rear calf (3) is connected via a flange shaft (14) to form a rotating pair; a pair of guide wheels (36) are respectively arranged on the front and rear sides of the chassis, and a laser ranging module (54) is arranged at the front of the bottom, and a positioning module (58) is arranged inside; The thigh (6) is connected to the rotor of the joint motor (11) through the frame (7) and the joint motor connector; the stator of the joint motor (11) is fixedly connected to the chassis through a fixing plate (12); the thigh (6) connected to the front calf (2) is raised by a raising plate (10); a leg limit block (8) is provided at the connection between the thigh (6) and the frame (7), wherein a limit raising plate (9) is installed at the thigh (6) connected to the front calf (2); Determine the wheel diameter range according to the initial speed and power requirements for hill climbing and the characteristic parameters of the hub motor to achieve matching; The control hardware is two single-chip microcomputers (46), and two central boards (45) are provided to distribute power for the circuits.

2. The design of a wheel-legged balancing robot capable of active jumping according to claim 1 is characterized in that: The main structure of the chassis comprises a "well"-shaped frame formed by transverse aluminum squares (18) and longitudinal aluminum squares (19), and a plate is used to cover and shield the robot line; a fixed aluminum square (13) and a hardware mounting plate (47) are fixedly connected to the "well"-shaped frame; the robot chassis is connected to the frame (7) through a transverse baffle (31) to form an upper enclosure plate, and the connection between the two plates is connected by a four-hole adapter (43); the lower enclosure plate is composed of two completely identical lower side plates (56) connected to the bottom side plate (48) through a four-hole adapter (43), forming a side enclosure; the robot is surrounded by four completely identical oblique anti-collision plates (44) at the front and back; four completely identical bottom center side plates (52) form the bottom enclosure, which are connected to the bottom center plate (49) through a four-hole adapter (43) to form a bottom enclosure structure.

3. The design of a wheel-legged balancing robot capable of active jumping according to claim 1 is characterized in that: The thigh (6) and the two shanks are connected by plug bolts (4), and two pairs of flange bearings (16) are arranged to be embedded in the shanks. The flange bearings bear radial forces, and a thrust bearing (15) is sandwiched between the pair of flange bearings (16) to bear axial forces. Two gaskets (5) are used to isolate the inner and outer rings of the flange bearings (16), and nuts (17) are screwed on to form a rotating pair.

4. The design of a wheel-legged balancing robot capable of active jumping according to claim 1 is characterized in that: Taking the moment of inertia into consideration, the length ratio of the thigh and calf is designed to be suitable for jumping, and the topology of the plate is optimized.

5. The design of a wheel-legged balancing robot capable of active jumping according to claim 1, characterized in that: In order to meet the requirements of the robot climbing over a slope, the peak torque T max The following relationship is satisfied between the wheel diameter of the wheel set (1): , where θ is the flight slope, m is the mass of the robot, and T max is the peak torque, R max is the maximum diameter of the wheel set, g is the gravity constant; there is the following relationship between the initial speed v and the wheel diameter: , where R min is the minimum diameter of the wheel set, n is the wheel speed; there is the following relationship between the slope power P and the motor torque T: ; Given θ, m, T max , v, P, and n, the diameter range of the wheelset can be solved.

6. The design of a wheel-legged balancing robot capable of active jumping according to claim 1 or 5, characterized in that: The wheel set (1) comprises a two-layer wheel rim (25) and a wheel hub motor (23), the outer ring of which is covered with rubber to form a rubber ring (27) and is clamped by an inner plate (28) and an outer plate (20); The inner ring is filled with a support member (26); the inner plate (28) and the outer plate (20) are connected by bolts through evenly distributed through holes on the wheel rim (25); the hub motor (23) is connected to the outer plate (20) through a pin coupling (22) and a load-reducing plate (21) having a diameter larger than a distribution circle diameter of the motor bolts to form a motor rotor; the motor stator is connected to the front lower leg (2) through a motor connector (24).

7. The design of a wheel-legged balancing robot capable of active jumping according to claim 1, characterized in that: The guide wheel (36) is sandwiched between the guide wheel outer mounting plate (34) and the guide wheel inner mounting plate (35); a space in the middle of the guide wheel is padded with a pad column (37); the guide wheel is fixed by mounting bolts (38) and mounting nuts (39); the guide wheel inner mounting plate (35) is inserted into the guide wheel upper support plate (32) and the guide wheel lower support plate (33) through a plate insertion structure; the guide wheel outer mounting plate is connected to the guide wheel upper support plate (32) and the guide wheel lower support plate (33) through a four-hole adapter (43).

8. The design of a wheel-legged balancing robot capable of active jumping according to claim 1, characterized in that: The joint motor connecting member comprises an inner ring plate (59) with a plurality of first through holes (591) formed on the ring surface, and an outer ring plate (60) with a plurality of second through holes (592) formed on the ring surface; the joint motor connecting member is externally connected to a cross roller bearing (61) and a bearing housing (62); the first through holes (591) are used to connect the motor rotor and the leg, respectively corresponding to the pin holes and bolt holes of the joint motor; and the second through holes (601) are used to connect the leg connecting rod.

9. A control method for a wheel-legged balancing robot capable of active jumping, characterized in that: The design of the wheel-legged balancing robot capable of active jumping according to any one of claims 1 to 8 comprises the following process: The robot system model is simplified to an inverted pendulum model, and the change in the robot leg length is ignored. The defined state vector includes variables related to the posture of the robot's upper structure and legs and the movement of the driving wheels; the variables include but are not limited to the body posture angle, the pendulum angle, and the angle of the line connecting the driving wheel shaft and the center of the motor shaft of the two joints of the leg relative to the inertial system; According to the above wheel-leg inverted pendulum model, the control law is designed as a linear combination of the system states, namely: ,in, is the state vector, is the control vector; is the feedback gain; The feedback gain is calculated using a linear quadratic regulator (LQR) and the reference input is added to the input of the linear quadratic regulator, that is: The reference input The robot position is expected constitute: The linear quadratic regulator is linearized every 10 mm within the leg length range, and its feedback gain matrix is ​​solved; a polynomial equation is fitted to each element of the matrix as the leg length changes, and the following is obtained: ； In the formula, is the gain matrix related to leg length, When L0=0, the matrix K ij The value of , which is the constant term of the polynomial, is the coefficient of the first-order derivative of the matrix elements as the leg length L0 changes, is the coefficient of the second-order derivative of the matrix elements as the leg length L0 changes, is the feedback gain matrix K ij The coefficient of the third-order derivative of the elements of with the leg length L0, For leg length; It is deduced that the robot longitudinal motion control law is: ； When the robot enters the active jumping mode, the laser ranging module 54 senses the obstacle ahead and automatically reproduces the jumping control code, and the four joint motors (11) work together to control the leg retraction-leg extension-leg retraction to complete the jumping action.

10. The control method of a wheel-legged balancing robot capable of active jumping according to claim 9, characterized in that: The inverted pendulum model adopts a segmented modeling approach, firstly establishing dynamic equations for the driving wheel, the pendulum rod and the body respectively, and then combining these equations to obtain a state space model by eliminating intermediate variables.

Images