Robot control methods, devices, robots, storage media, and program products

Abstract

This application discloses a robot control method, device, robot, storage medium, and program product, relating to the field of artificial intelligence technology. The method includes: adjusting the tilt angle between the torso and the supporting surface to switch from a wheeled standing state to a combined legged and wheeled prone state; in the wheeled standing state, the robot maintains its standing position solely through its drive wheels; swinging the torso in a first direction with the foot corresponding to the second mechanical leg group as the center, controlling the drive wheel corresponding to the first mechanical leg group to detach from the supporting surface; swinging the torso in a second direction with the foot corresponding to the second mechanical leg group as the center, controlling the foot corresponding to the first mechanical leg group to touch the ground, entering a legged prone state; in the legged prone state, the robot maintains its prone position solely through its feet, with the first direction and the second direction being opposite. This application solves the problem of robots being unable to switch from standing to prone due to limited footholds, expanding the robot's applicable scenarios and improving its ability to automatically switch states.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, and in particular to a robot control method, device, robot, storage medium, and program product. Background Technology

[0002] With the development of artificial intelligence technology, some organizations and research institutions have launched a number of robots that can switch states automatically, such as switching from standing to crawling or from standing to lying down.

[0003] Taking quadruped robots as an example, most current research on the autonomous state switching of quadruped robots is conducted within a sufficiently large foothold. For some special scenarios, there is relatively little research on the autonomous state switching of quadruped robots, resulting in insufficient autonomous state switching capabilities of current quadruped robots in some special scenarios (such as narrow footholds). Summary of the Invention

[0004] This application provides a robot control method, device, robot, storage medium, and program product, which can improve the robot's ability to automatically switch states, thereby improving the robot's applicability. The technical solution is as follows:

[0005] According to one aspect of the embodiments of this application, a robot control method is provided. The robot includes a torso, and a first mechanical leg assembly and a second mechanical leg assembly connected to the torso. The first mechanical leg assembly includes at least two wheel-driven mechanical legs, which move via drive wheels or feet. The method includes:

[0006] Adjust the tilt angle between the torso and the support surface to switch from a wheeled standing state to a foot-wheeled compound prone state; wherein, in the wheeled standing state, the robot maintains its standing position solely through the drive wheels corresponding to the first mechanical leg group; in the foot-wheeled compound prone state, the robot maintains its prone position through the drive wheels corresponding to the first mechanical leg group and the feet corresponding to the second mechanical leg group;

[0007] The torso swings in the first direction with the foot corresponding to the second mechanical leg group as the center, and controls the drive wheel corresponding to the first mechanical leg group to disengage from the support surface;

[0008] The torso swings in a second direction with the foot corresponding to the second mechanical leg group as the center, controlling the foot corresponding to the first mechanical leg group to land and enter a footed prone state; wherein, in the footed prone state, the robot maintains its prone position only through its feet;

[0009] Wherein, the first direction is opposite to the second direction.

[0010] According to one aspect of the embodiments of this application, a robot control method is provided. The robot includes a torso, and a first mechanical leg assembly and a second mechanical leg assembly connected to the torso. The first mechanical leg assembly includes at least two wheel-driven mechanical legs, which move via drive wheels or feet. The method includes:

[0011] Adjust the tilt angle between the torso and the support surface to switch from a wheeled standing state to a foot-wheeled compound prone state; wherein, in the wheeled standing state, the robot maintains its standing position solely through the drive wheels corresponding to the first mechanical leg group; in the foot-wheeled compound prone state, the robot maintains its prone position through the drive wheels corresponding to the first mechanical leg group and the feet corresponding to the second mechanical leg group;

[0012] Control the torso to tilt towards the first mechanical leg in the first mechanical leg group. When the second mechanical leg in the first mechanical leg group detaches from the support surface under the action of gravity, switch the second mechanical leg from the drive wheel ground support state to the foot ground support state.

[0013] The torso is tilted toward the second mechanical leg. When the first mechanical leg detaches from the support surface under the action of gravity, the first mechanical leg is switched from the state where the drive wheel is on the ground to the state where the foot is on the ground.

[0014] According to one aspect of the embodiments of this application, a control device for a robot is provided. The robot includes a torso, and a first mechanical leg assembly and a second mechanical leg assembly connected to the torso. The first mechanical leg assembly includes at least two foot-wheel composite mechanical legs, which move via drive wheels or feet. The device includes:

[0015] The wheel-mounted prone switching module is used to adjust the tilt angle between the torso and the support surface, switching from a wheeled standing state to a wheeled compound prone state. In the wheeled standing state, the robot maintains its standing position solely through the drive wheels corresponding to the first mechanical leg group; in the wheeled compound prone state, the robot maintains its prone position through the drive wheels corresponding to the first mechanical leg group and the feet corresponding to the second mechanical leg group.

[0016] The drive wheel lift-off module is used to swing the torso in a first direction with the foot corresponding to the second mechanical leg group as the center, and control the drive wheel corresponding to the first mechanical leg group to detach from the support surface.

[0017] The foot-prone switching module is used to swing the torso in a second direction with the foot corresponding to the second mechanical leg group as the center, and control the foot corresponding to the first mechanical leg group to land and enter a foot-prone state; wherein, in the foot-prone state, the robot only maintains a prone position through its feet; wherein, the first direction is opposite to the second direction.

[0018] According to one aspect of the embodiments of this application, a control device for a robot is provided. The robot includes a torso, and a first mechanical leg assembly and a second mechanical leg assembly connected to the torso. The first mechanical leg assembly includes at least two foot-wheel composite mechanical legs, which move via drive wheels or feet. The device includes:

[0019] The wheel-mounted prone switching module is used to adjust the tilt angle between the torso and the support surface, switching from a wheeled standing state to a wheeled compound prone state. In the wheeled standing state, the robot maintains its standing position solely through the drive wheels corresponding to the first mechanical leg group; in the wheeled compound prone state, the robot maintains its prone position through the drive wheels corresponding to the first mechanical leg group and the feet corresponding to the second mechanical leg group.

[0020] The first foot switching module is used to control the torso to tilt towards the first mechanical leg in the first mechanical leg group. When the second mechanical leg in the first mechanical leg group leaves the support surface under the action of gravity, the second mechanical leg is switched from the drive wheel ground support state to the foot ground support state.

[0021] The second foot switching module is used to control the torso to tilt towards the second mechanical leg, and to switch the first mechanical leg from the drive wheel ground support state to the foot ground support state when the first mechanical leg detaches from the support surface under the action of gravity.

[0022] According to one aspect of the embodiments of this application, a robot is provided, the robot including a processor and a memory, the memory storing a computer program, the computer program being loaded and executed by the processor to implement the control method of the robot described above.

[0023] According to one aspect of the embodiments of this application, a computer-readable storage medium is provided, wherein a computer program is stored in the storage medium, the computer program being loaded and executed by a processor to implement the above-described robot control method.

[0024] According to one aspect of the embodiments of this application, a computer program product or computer program is provided, which includes computer instructions stored in a computer-readable storage medium. A robot's processor reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the robot to perform the aforementioned robot control method.

[0025] The technical solutions provided in this application embodiment may have the following beneficial effects:

[0026] By adjusting the tilt angle between the torso and the support surface, the robot can switch from a wheeled standing state to a footed lying state. Then, based on the second mechanical leg assembly, the torso is swung, and the drive wheels corresponding to the first mechanical leg assembly are controlled to detach from the support surface. During the swinging process, the drive wheels are switched from landing support to foot support, thereby achieving the switch from a wheeled standing state to a footed lying state. This increases the robot's self-switching method and improves the robot's self-switching flexibility.

[0027] In addition, by adopting the technical solution provided in the embodiments of this application, the robot can switch from a wheeled standing state to a footed lying state in some special scenarios (such as a narrow footing area), which solves the problem that the robot cannot switch from a wheeled standing state to a footed lying state due to special scenarios. This expands the applicable scenarios of the robot and improves the robot's ability to switch states automatically, thereby improving the applicability of the robot.

[0028] In addition, for the wheel-wheel hybrid quadruped robot, the technical solution provided in the embodiments of this application enables the wheel-wheel hybrid quadruped robot to autonomously switch between wheeled standing state and legged lying state even in situations where the footing range is narrow (such as plum blossom stakes), thereby improving the applicability of the wheel-wheel hybrid quadruped robot. Attached Figure Description

[0029] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0030] Figure 1 This is a schematic diagram of a wheel-driven quadruped robot provided in one embodiment of this application;

[0031] Figure 2 This is a schematic diagram of a foot-wheel composite mechanical leg with a drive wheel motion mode provided in one embodiment of this application;

[0032] Figure 3 This is a schematic diagram of a foot-wheel composite mechanical leg according to an embodiment of this application, which provides a foot movement mode;

[0033] Figure 4 This is a flowchart of a robot control method provided in one embodiment of this application;

[0034] Figure 5 This is a schematic diagram of a robot in a wheeled standing state according to an embodiment of this application;

[0035] Figure 6 This is a schematic diagram of a robot in a compound prone position with casters provided in one embodiment of this application;

[0036] Figure 7 This is a schematic diagram of a robot during a falling process according to one embodiment of this application;

[0037] Figure 8 This is a schematic diagram of a robot in a swinging, airborne state according to an embodiment of this application;

[0038] Figure 9 This is a schematic diagram of a robot in a desired swing posture according to an embodiment of this application;

[0039] Figure 10 This is a schematic diagram of a robot in a swing-ready state according to an embodiment of this application;

[0040] Figure 11 This is a schematic diagram of a robot in a swing termination desired posture according to an embodiment of this application;

[0041] Figure 12 This is a schematic diagram of a robot in a legged prone position according to an embodiment of this application;

[0042] Figure 13 This is a schematic diagram of a robot located on a pile surface according to one embodiment of this application;

[0043] Figure 14 This is a flowchart illustrating how a robot switches from a wheeled standing state to a legged lying state, according to one embodiment of this application.

[0044] Figure 15 This is a schematic diagram of a wheeled quadruped robot switching from a wheeled standing state to a legged crouching state according to an embodiment of this application;

[0045] Figure 16 This is a flowchart of a robot control method provided in another embodiment of this application;

[0046] Figure 17This is a schematic diagram of a wheel-driven quadruped robot switching from a wheeled standing state to a legged crouching state, according to another embodiment of this application.

[0047] Figure 18 This is a flowchart of another embodiment of the present application showing the robot switching from a wheeled standing state to a legged lying state;

[0048] Figure 19 This is a schematic diagram of a method for obtaining the desired landing posture provided in one embodiment of this application;

[0049] Figure 20 This is a schematic diagram of a robot control strategy provided in one embodiment of this application;

[0050] Figure 21 This is a block diagram of a robot control device provided in one embodiment of this application;

[0051] Figure 22 This is a block diagram of a robot control device provided in another embodiment of this application;

[0052] Figure 23 This is a simplified structural block diagram of a robot provided in one embodiment of this application. Detailed Implementation

[0053] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0054] Artificial intelligence (AI) is the theory, methods, technology, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results. In other words, AI is a comprehensive technology within computer science that attempts to understand the essence of intelligence and produce a new kind of intelligent machine that can react in a way similar to human intelligence. AI studies the design principles and implementation methods of various intelligent machines, enabling them to possess the functions of perception, reasoning, and decision-making.

[0055] Artificial intelligence (AI) is a comprehensive discipline encompassing a wide range of fields, including both hardware and software technologies. Fundamental AI technologies generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies primarily include computer vision, speech processing, natural language processing, and machine learning / deep learning.

[0056] With the research and advancement of artificial intelligence (AI) technology, AI is being studied and applied in various fields, such as smart homes, smart wearable devices, virtual assistants, smart speakers, smart marketing, autonomous driving, drones, robots, smart healthcare, and smart customer service. It is believed that with the development of technology, AI will be applied in more fields and play an increasingly important role.

[0057] This application's technical solution primarily relates to robotics within artificial intelligence, specifically intelligent robot control. A robot is a mechatronic electronic device that combines mechanical transmission and modern microelectronics technology to mimic certain human skills. Robots have evolved based on electronic, mechanical, and information technologies. A robot doesn't necessarily have to resemble a human; as long as it can autonomously complete tasks and commands assigned by humans, it belongs to the robot family. A robot is an automated machine possessing some intelligent abilities similar to humans or other living organisms, such as perception, planning, movement, and coordination. It is a highly flexible automated machine. With the development of computer and artificial intelligence technologies, robots have seen significant improvements in both functionality and technology. Mobile robots and technologies such as robot vision and touch are typical examples.

[0058] The method provided in this application can be executed by a computer device, which refers to an electronic device with data computing, processing, and storage capabilities. This computer device can be a terminal such as a PC (Personal Computer), tablet computer, smartphone, desktop computer, or intelligent robot; or it can be a server. The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0059] The robot control method provided in this application enables the robot to autonomously switch between a wheeled standing state and a legged prone state even when the landing area is small. For example, using the robot control method provided in this application, the robot can autonomously switch between a wheeled standing state and a legged prone state on plum blossom stakes.

[0060] In one example, consider a quadruped robot with both legs and wheels. A quadruped robot with both legs and wheels is defined as a quadruped robot with two front legs and two hind legs, and wheels are mounted at the knee joints, allowing the robot to stand on all four legs, on two wheels, or in a combination of both. These wheels can be mounted only on the hind legs, only on the front legs, or on both front and hind legs.

[0061] like Figure 1 As shown, an exemplary schematic diagram of a foot-wheel hybrid quadruped robot is illustrated. This foot-wheel hybrid quadruped robot may include: a camera 101, a torso 102, two front legs 103 and 104 (left and right), and two hind legs 105 and 106 (left and right), each with three degrees of freedom. The front legs 103, 104, and hind legs 105 and 106 are connected to the torso 102. Figure 1 Sub-figure (a) is a schematic diagram of the lateral structure of the wheel-driven quadruped robot. Figure 1 Sub-figure (b) is a schematic diagram of the oblique forward structure of the wheel-driven quadruped robot.

[0062] The front legs 103 and 104 form the second mechanical leg assembly of the foot-wheel composite quadruped robot. Front wheels 107 and 108 are respectively installed at the knee joints of the front legs 103 and 104. The front wheels 107 and 108 can be either passive, undriven wheels (i.e., driven wheels) or driven wheels; this embodiment does not limit the specific type of wheel. In some feasible embodiments, the front legs 103 and 104 may not have front wheels.

[0063] The hind legs 105 and 106 form the first mechanical leg group of the foot-wheel composite quadruped robot. The hind legs 105 and 106 are respectively equipped with rear wheels 109 and 110 at their knee joints. The rear wheels 109 and 110 can be driven wheels (i.e., active wheels).

[0064] refer to Figure 2Each leg of the foot-wheel quadruped robot is equipped with three joint motors: a trunk joint motor 111, a hip joint motor 112, and a knee joint motor 113. The trunk joint motor 111, located at the connection between the trunk and the leg, drives the entire leg to rotate laterally; this motor can be an ab / ad motor. The hip joint motor 112, located at the end of the leg connected to the trunk joint motor 111 (i.e., the hip joint of the leg), drives the thigh to rotate. The knee joint motor 113, located at the end of the thigh connected to the hip joint motor 112, drives the lower leg to rotate. A pulley 118 on the knee joint motor 113 and a pulley 119 at the knee joint are connected by a belt, allowing the knee joint motor 113 to control the rotation of the lower leg via the belt.

[0065] Unlike the front legs, the lower legs of the hind legs are equipped with linear actuators 114 and steel blocks 115, and the thighs of the hind legs are equipped with magnets 116. Correspondingly, the drive wheels are equipped with grooved plates 117.

[0066] Linear driver 114 can be a linear motor, see reference. Figure 2 When the linear actuator 114 extends, its protruding portion inserts into the slot plate 117. Thus, while the knee joint motor 113 drives the drive wheel via the belt, it also drives the rotation of the lower leg, i.e., the hind leg enters a foot-movement mode. (Reference) Figure 3 When the linear actuator 114 retracts, its extended portion disengages from the slot plate 117. Therefore, while the knee joint motor 113 drives the drive wheel via the belt, the connection between the lower leg and the drive wheel is severed. Consequently, the knee joint motor 113 only drives the drive wheel to rotate, unable to drive the lower leg to rotate; that is, the hind leg enters the drive wheel movement mode. Furthermore, refer to... Figure 3 In the drive wheel motion mode, the lower leg is fixed to the thigh by the attraction of the steel block 115 and the magnet 116. With this structural design, the hind leg can be autonomously switched between foot motion mode and drive wheel motion mode by controlling the linear actuator 114.

[0067] The camera 101 has image acquisition capabilities, such as acquiring images of the surrounding environment, so as to analyze the images of the surrounding environment and plan a suitable movement path for the wheel-driven quadruped robot.

[0068] In the embodiments of this application, since each hind leg of the footwheel-driven quadruped robot is a footwheel-driven mechanical leg, which can move through the drive wheel or the foot, the footwheel-driven quadruped robot can have a variety of motion states, such as a two-wheel balance state, a four-legged prone state, a four-wheel prone state, and a footwheel-driven prone state.

[0069] The dual-wheel balance state refers to the state in which the torso of the quadruped robot is vertical, the front legs are suspended in the air, and the rear wheels 109 and 110 are in contact with the support surface (such as the ground, pile surface, etc.). The robot moves back and forth in front of the support surface to maintain balance and realize actions such as standing on dual wheels and gliding on dual wheels.

[0070] The quadrupedal lying state refers to the use of four legs of the quadruped robot to contact the supporting surface, and the joint motors drive the four legs to achieve actions such as quadrupedal lying and quadrupedal crawling.

[0071] The four-wheeled prone state refers to the quadruped robot using its front wheels 107, 108, rear wheels 109, and 110 to contact the support surface, while other parts are away from the support surface, enabling it to perform actions such as standing on four wheels and gliding on four wheels.

[0072] The compound crawling state refers to the compound crawling quadruped robot using its front legs 103 and 104 to contact the support surface, and its rear wheels 109 and 110 to contact the support surface, thus achieving the compound crawling action.

[0073] The structure of the robot in the embodiments of this application has been described above. The control method of the robot will be described below.

[0074] Please refer to Figure 4 The diagram shows a flowchart of a robot control method provided in one embodiment of this application. The execution subject of each step of the method can be a robot (such as the above-mentioned wheeled quadruped robot). The method can include the following steps (401-403).

[0075] Step 401: Adjust the tilt angle between the torso and the support surface to switch from the wheeled standing state to the foot-wheeled compound prone state; wherein, in the wheeled standing state, the robot maintains standing only through the drive wheels corresponding to the first mechanical leg group; in the foot-wheeled compound prone state, the robot maintains prone through the drive wheels corresponding to the first mechanical leg group and the feet corresponding to the second mechanical leg group.

[0076] The robot may include a torso, and a first mechanical leg assembly and a second mechanical leg assembly connected to the torso. The first mechanical leg assembly includes at least two wheel-driven mechanical legs, which move via drive wheels or feet. The second mechanical leg assembly may include at least two ordinary mechanical legs, or at least two wheel-driven mechanical legs. The robot in this embodiment is the same as described in the above embodiments. Content not described in this embodiment can be referred to the above embodiments, and will not be repeated here.

[0077] A coordinate system is constructed with the landing point between the drive wheel and the support surface as the origin. When the horizontal axis of the coordinate system is parallel to the support surface, the above tilt angle can refer to the angle between the robot's torso and the horizontal axis in the counterclockwise direction.

[0078] The wheeled standing state refers to a robot maintaining a longitudinal posture with its torso in contact with the support surface through the drive wheels corresponding to the first mechanical leg group, while other parts of the robot remain in contact with the support surface. The foot-wheel compound prone state refers to a robot maintaining a lateral posture with its torso in contact with the support surface through the drive wheels corresponding to the first mechanical leg group and the feet corresponding to the second mechanical leg group. Both the first and second mechanical leg groups consist of mechanical legs on the same side. Taking a foot-wheel compound quadruped robot as an example, the mechanical legs in the first mechanical leg group can be composed of the robot's two hind mechanical legs, and the mechanical legs in the second mechanical leg group can be composed of the robot's two front mechanical legs.

[0079] For example, refer to Figure 5 Robot 501 (a quadruped robot with a wheel) is in a wheeled standing state (specifically corresponding to a two-wheeled balance state). Robot 501 is kept standing on the ground (i.e., the support surface) by the drive wheels on its two rear mechanical legs. The torso of Robot 501 is vertically upward, and it can also have a certain tilt angle. The two front mechanical legs of the robot are suspended in the air.

[0080] Optionally, in the wheeled standing state, the mechanical legs in the second mechanical leg assembly can have any posture, and this embodiment of the application does not limit this. In the wheeled standing state and the combined foot-wheel prone state, the mechanical legs in the first mechanical leg assembly are in drive wheel motion mode.

[0081] For example, refer to Figure 6 Robot 501 is in a foot-wheel compound prone position. The drive wheels on the two rear mechanical legs of robot 601 are in contact with the ground, and the feet of the two front mechanical legs are in contact with the ground, so that robot 501 remains in a prone position.

[0082] In one example, the process of switching from a wheeled standing position to a combined wheeled and prone position can be as follows:

[0083] 1. Obtain the initial posture corresponding to the wheeled standing state. This initial posture includes the initial angles of each joint motor of the robot in the wheeled standing state.

[0084] The initial posture corresponding to the wheeled standing state can refer to the posture at the moment when the robot begins to switch states. This initial posture may also include the initial angle between the mechanical leg in the first mechanical leg group and the supporting surface, and the initial tilt angle between the torso and the supporting surface.

[0085] refer to Figure 5In the initial posture, robot 501 stands upright. The initial angle between the torso and each thigh of robot 501 is 0 degrees (i.e., the initial angle corresponding to the hip joint is 0 degrees). The lower leg of the rear mechanical leg is fixed on the thigh (the corresponding initial angle does not need to be considered). The initial angle corresponding to the lower leg of the front mechanical leg (i.e., the angle corresponding to the knee joint) is not limited.

[0086] 2. Obtain the expected landing posture corresponding to the foot-wheel compound prone state. The expected landing posture includes the expected landing angle of each joint motor in the foot-wheel compound prone state, as well as the expected landing tilt angle between the torso and the support surface.

[0087] The expected landing posture corresponding to the foot-wheel compound prone state can refer to the expected posture of the foot of the mechanical leg in the second mechanical leg group when it just lands, or it can refer to the expected posture after the robot has landed and stabilized.

[0088] For example, refer to Figure 6 Under the desired landing posture, the desired landing tilt angle between the torso of robot 501 and the ground can be set and adjusted based on empirical values ​​(e.g., greater than or equal to 0 degrees). Since the lengths of each structure corresponding to robot 501, as well as the distances between the landing points of the front and rear mechanical legs, are fixed, and the desired tilt angles can be set individually, the desired landing angles corresponding to each joint of robot 501 can be calculated geometrically (which can be further converted into the desired landing angles of the joint motors). This will be explained in detail below and will not be repeated here. In addition, since the angle between the lower leg and thigh of the rear mechanical leg is fixed (e.g., 30 degrees), the desired landing angle corresponding to the knee joint of the rear mechanical leg does not need to be calculated separately.

[0089] Optionally, when the robot lands, the robot's torso can be set to horizontal, i.e., the desired tilt angle upon landing is 0 degrees. Alternatively, the torso can be adjusted to horizontal or left unadjusted after the robot lands.

[0090] 3. Based on the initial posture and the expected landing posture, obtain the landing planning posture sequence corresponding to the robot. This landing planning posture sequence includes the expected posture of the robot at each moment during the landing phase.

[0091] Optionally, the planned landing attitude sequence can be obtained by interpolating along the time dimension, starting from the initial attitude and ending with the desired landing attitude. Alternatively, the planned landing attitude sequence can be obtained by linearly transforming the initial attitude and ending with the desired landing attitude.

[0092] 4. Based on the landing posture sequence, adjust the tilt angle between the torso and the support surface to control the robot's center of gravity to move in the first direction.

[0093] The first direction can be any orientation corresponding to the torso. For example, refer to... Figure 5 The first direction can refer to the front of robot 501 or the rear of robot 501.

[0094] In one example, based on the planned landing posture sequence, the drive wheels corresponding to the first mechanical leg assembly can be controlled to move in the second direction. The tilt angle between the torso and the support surface increases due to inertia, causing the robot's center of gravity to move in the first direction. The second direction is opposite to the first direction.

[0095] For example, refer to Figure 7 After the robot 501 controls the drive wheels on the mechanical legs to slide backward, the torso tilts forward under the action of inertia, thereby causing the center of gravity to move forward.

[0096] In another example, the angle between the first mechanical leg assembly and the torso can be adjusted directly in the first direction, causing the robot's torso to tilt in the first direction, thereby causing the robot's center of gravity to move in the first direction.

[0097] For example, or refer to Figure 7 Robot 501 keeps the drive wheels on its rear mechanical legs stationary and adjusts the angle between the rear mechanical legs and the torso, causing the torso to tilt forward and thus shifting the center of gravity forward.

[0098] Optionally, if the robot's torso is already tilted in the first direction in the initial state, then the robot does not need to control the center of gravity to move in the first direction.

[0099] 5. With the robot's center of gravity in a position that allows for torso descent, adjust the angles of the motors at each joint according to the planned landing posture sequence, and enter a compound prone position upon landing.

[0100] Optionally, refer to Figure 7 When the center of gravity of robot 501 exceeds the rear mechanical legs, the torso of robot 501 begins to fall under the action of gravity. During the fall of the torso, the angles of the corresponding joint motors of robot 501 can be adjusted according to the landing posture sequence, so that the joint motors of robot 501 gradually rotate to the desired landing angle and the robot 501 gradually switches to the desired landing posture.

[0101] Optionally, the initial posture and the desired landing posture can be obtained by the designer and then input into the robot (or the robot's control system), or the robot can obtain them automatically. The landing planning posture sequence can be calculated by the robot based on the initial posture and the desired landing posture, or it can be calculated by the designer based on the initial posture and the desired landing posture and then input into the robot. This embodiment of the application does not limit this.

[0102] Step 402: Swing the torso in the first direction with the foot corresponding to the second mechanical leg group as the center, and control the drive wheel corresponding to the first mechanical leg group to disengage from the support surface.

[0103] Optionally, the torso can be swung in the first direction with the feet of each of the second mechanical legs as the center, or the torso can be swung in the first direction with the line connecting the feet of each of the second mechanical legs as the axis. This embodiment does not limit the specific method used. The feet can also be replaced by the landing point between the feet and the support surface.

[0104] For example, refer to Figure 8 Robot 501 swings its torso forward, using the first landing point between its left front mechanical leg and the ground as the first center and the second landing point between its right front mechanical leg and the ground as the second center, so that the drive wheels corresponding to its two rear mechanical legs lift off the ground. Alternatively, robot 501 swings its torso forward, using the line connecting the first and second landing points as the axis, so that the drive wheels corresponding to its two rear mechanical legs lift off the ground.

[0105] In one example, step 502 may also include the following:

[0106] 1. Maintain the robot's footing point, move the torso in the first direction, and switch from the foot-wheel compound prone state to the swing preparation state; wherein, in the swing preparation state, the robot has a target angular velocity that causes the drive wheel corresponding to the first mechanical leg group to detach from the support surface.

[0107] In this embodiment, the swing preparation state refers to the robot's preparation state for swinging its torso. After the robot completes the swing preparation process, it has a target angular velocity that causes the drive wheel corresponding to the first mechanical leg assembly to disengage from the support surface. The target angular velocity can be set and adjusted based on empirical values. Optionally, when acquiring the target angular velocity, the robot's center of gravity does not exceed the second mechanical leg assembly to prevent the robot from tipping over in the first direction.

[0108] For example, the process of switching from the wheel-assisted ground position to the swing preparation position can be as follows:

[0109] Obtain the expected swing posture of the robot, which includes the expected swing angle between the mechanical leg in the first mechanical leg group and the support surface, the expected swing angle of each joint motor of the robot in the swing preparation state, and the expected swing tilt angle between the torso and the support surface.

[0110] The desired swing posture can refer to the critical state where the drive wheel corresponding to the first mechanical leg assembly is about to detach from the support surface. The desired swing angle, desired swing angle, and desired swing tilt angle can be set and adjusted based on empirical values. Under the combined effect of the desired swing angle, desired swing angle, and desired swing tilt angle, the robot can achieve the target angular velocity.

[0111] For example, refer to Figure 9 In the desired swing posture, the angle between the two rear mechanical legs of robot 501 and the ground satisfies the desired swing angle, the angles corresponding to the joint motors of robot 501 satisfy their respective desired swing angles, and the tilt angle between the torso of robot 501 and the ground satisfies the desired swing tilt angle. At this time, robot 501 has a target angular velocity. The angles corresponding to the joint motors can be converted to the corresponding joint angles; for example, the angle corresponding to the knee joint motor can be converted to the angle corresponding to the knee joint.

[0112] Based on the posture corresponding to the compound ground position of the footwheel and the expected posture for starting the swing, the corresponding swing preparation posture sequence of the robot is determined; wherein, the swing preparation posture sequence includes the expected posture of the robot at each moment in the swing preparation stage, and the expected angle between the mechanical leg in the first mechanical leg group and the support surface at each moment in the swing preparation stage.

[0113] Optionally, the preparatory posture sequence for the swing can be obtained by interpolating along the time dimension, starting from the posture corresponding to the compound ground position of the foot wheel and ending with the desired swing posture. Alternatively, the preparatory posture sequence for the swing can be obtained by linearly transforming the posture corresponding to the compound ground position of the foot wheel and ending with the desired swing posture.

[0114] The postures corresponding to the wheeled compound ground position and the expected swing posture can be obtained by the designer and then input into the robot, or the robot can obtain them automatically. The landing planning posture sequence can be calculated by the robot based on the postures corresponding to the wheeled compound ground position and the expected swing posture, or it can be calculated by the designer based on the wheeled compound ground position and the expected swing posture, and then input into the robot. This application embodiment does not limit this.

[0115] Based on the swing preparation posture sequence, maintain the robot's landing point and adjust the angle between the mechanical legs in the first mechanical leg group and the supporting surface, as well as the angles of each joint motor, to control the torso to move in the first direction. For example, refer to... Figure 6 and Figure 9 Robot 501, following the preparatory posture sequence, keeps its various landing points stationary and gradually adjusts the angles of the motors at each joint to move its torso forward.

[0116] Once the robot's angular velocity meets the target angular velocity, it enters the swing preparation state. For example, refer to... Figure 9 Once the robot 501's posture is adjusted to the desired swing posture, the robot 501 has the target angular velocity, and the robot 501 completes the swing preparation process.

[0117] This application does not limit the timing of the first mechanical leg assembly switching from drive wheel motion to foot motion, as long as it ensures that the feet of the first mechanical leg assembly meet the landing requirements before the robot enters the footed lying state. For example, the linear actuator corresponding to the first mechanical leg assembly can be adjusted before the swing preparation process to control the first mechanical leg assembly to switch from drive wheel motion to foot motion; the switch can also be controlled during the swing preparation process; or the switch can be controlled after the drive wheels have disengaged from the support surface.

[0118] 2. Driven by the inertia corresponding to the target angular velocity, the torso is controlled to swing in the first direction with the foot corresponding to the second mechanical leg group as the center, and the drive wheel corresponding to the first mechanical leg group is disengaged from the support surface.

[0119] For example, refer to Figure 10 and Figure 8 In the swing preparation state, robot 501 has an angular velocity ω. When the angular velocity ω meets the target angular velocity, the inertia corresponding to the angular velocity ω is sufficient to pull up the drive wheels corresponding to the two rear mechanical legs of robot 501, thereby causing the drive wheels to leave the ground.

[0120] Step 403: Swing the torso in the second direction with the foot corresponding to the second mechanical leg group as the center, and control the foot corresponding to the first mechanical leg group to land, entering a footed prone state; wherein, in the footed prone state, the robot only maintains its prone position through its feet; wherein, the first direction is opposite to the second direction.

[0121] Optionally, the foot landing point of the foot corresponding to the first mechanical leg group may be the same as or different from the foot landing point of the drive wheel corresponding to the first mechanical leg group, and the foot landing range of the foot corresponding to the first mechanical leg group may be the same as or different from the foot landing range of the drive wheel corresponding to the first mechanical leg group. This application embodiment does not limit this.

[0122] A legged prone position refers to a state in which a robot maintains a prone position by controlling the feet corresponding to the first and second robotic leg groups to contact the supporting surface. For example, see reference... Figure 11 Robot 501 keeps itself in a prone position by having its feet in contact with the ground.

[0123] In one example, after the drive wheel corresponding to the first mechanical leg group leaves the support surface, that is, during the process of the drive wheel corresponding to the first mechanical leg group being airborne, the linear actuator corresponding to the first mechanical leg group can be adjusted to control the first mechanical leg group to switch from drive wheel movement mode to foot movement mode.

[0124] Then, using the foot corresponding to the second mechanical leg assembly as the center, adjust the angle of the joint motor corresponding to the second mechanical leg assembly to swing the torso in the second direction. For example, refer to... Figure 8 and Figure 11 Robot 501 adjusts its two front mechanical legs and swings its torso backward, so that after the torso counteracts the inertia corresponding to the target angular velocity, it has a backward angular velocity. This backward angular velocity can be set and adjusted according to empirical values. Under the inertial effect of the backward angular velocity, the robot's torso begins to swing backward.

[0125] Based on the robot's desired swing termination posture, the angles of the motors of each joint of the robot are adjusted to control the foot of the first mechanical leg group to land; wherein, the desired swing termination posture includes the desired termination joint angle of the foot of the first mechanical leg group.

[0126] The swing termination desired posture refers to the robot's desired posture when the foot corresponding to the first robotic leg assembly just lands. The desired termination joint angle of the foot corresponding to the first robotic leg assembly is used to ensure that the foot of the first robotic leg assembly can contact the support surface. This desired termination joint angle can refer to the desired termination joint angle of the knee joint of the robotic leg in the first robotic leg assembly.

[0127] Optionally, after the robot's torso swings in the second direction, the angles of the corresponding joint motors can be adjusted according to the desired termination posture, so that the robot's posture gradually switches to the desired termination posture, allowing the feet of the first mechanical leg assembly to land on the ground. Specifically, during the swing of the robot's torso in the second direction, the inertia of the angular velocity in that direction gradually disappears, and the robot's torso begins to fall under the influence of gravity, causing the feet of the first mechanical leg assembly, which have already been adjusted to the desired termination posture, to land on the support surface.

[0128] The robot enters a legged prone position. Optionally, the robot can adjust from the desired posture after the swing terminates to the posture required for the legged prone position, thereby allowing the robot to smoothly enter the legged prone position. Optionally, after entering the legged prone position, the robot can further level its torso.

[0129] For example, refer to Figure 11Under the desired termination posture, the angles of the joint motors of robot 501 satisfy the desired termination joint angles corresponding to the desired termination posture, the angles between the feet of the two rear mechanical legs and the ground, and the angles between the feet of the two front mechanical legs and the ground satisfy the desired termination angles, and the tilt angle between the torso of robot 501 and the ground satisfies the desired termination tilt angles corresponding to the desired termination posture.

[0130] refer to Figure 12 Then, robot 501 makes minor adjustments to its posture to enter a legged prone position. Next, robot 501 levels its torso. If robot 501's torso is already horizontal in the legged prone position, then robot 501 does not need to level its torso again.

[0131] In one example, the landing range of the mechanical leg in the first mechanical leg group is smaller than a threshold range. Within this threshold range, the drive wheel and foot of the mechanical leg in the first mechanical leg group cannot land simultaneously. For example, refer to... Figure 13 The landing area of ​​the rear mechanical leg of robot 501 is a small pile surface, which is too small (e.g., the length of the pile surface is less than the length of the lower leg of the rear mechanical leg) to support the simultaneous landing of the drive wheel and foot of the rear mechanical leg. For example, the diameter of the pile surface corresponding to the rear mechanical leg is only 20cm (the diameter of the pile surface is less than the length of the lower leg of the rear mechanical leg), and the diameter of the pile surface corresponding to the front mechanical leg is only 12cm. Because the area of ​​the pile surface corresponding to the rear mechanical leg is not large enough to support the simultaneous contact between the drive wheel and foot of the rear mechanical leg and the pile surface, that is, the rear mechanical leg can only make point contact with the pile surface (e.g., the foot or the drive wheel). Using the conventional method (i.e., the foot and the drive wheel land simultaneously, and then the drive wheel is controlled to lift off the ground), the robot cannot switch from a wheeled standing state to a footed lying state.

[0132] In this case, the technical solution provided in the embodiments of this application can still complete the switch from wheeled standing state to legged lying state, thereby expanding the applicable scenarios of the robot and improving the robot's ability to switch states automatically.

[0133] In summary, the technical solution provided in this application, by adjusting the tilt angle between the torso and the support surface, achieves the switching from a wheeled standing state to a foot-wheeled compound prone state. Then, based on the second mechanical leg assembly, the torso is swung, controlling the drive wheel corresponding to the first mechanical leg assembly to detach from the support surface. During the swinging process, the drive wheel's ground support is switched to the foot's ground support, thereby achieving the switching from a wheeled standing state to a foot-wheeled prone state. This increases the robot's self-switching state method and improves the robot's self-switching state flexibility.

[0134] In addition, by adopting the technical solution provided in the embodiments of this application, the robot can switch from a wheeled standing state to a footed lying state in some special scenarios (such as a narrow footing area), which solves the problem that the robot cannot switch from a wheeled standing state to a footed lying state due to special scenarios. This expands the applicable scenarios of the robot and improves the robot's ability to switch states automatically, thereby improving the applicability of the robot.

[0135] In addition, for the wheel-wheel hybrid quadruped robot, the technical solution provided in the embodiments of this application enables the wheel-wheel hybrid quadruped robot to autonomously switch between wheeled standing state and legged lying state even in situations where the footing range is narrow (such as plum blossom stakes), thereby improving the applicability of the wheel-wheel hybrid quadruped robot.

[0136] In one exemplary embodiment, reference Figure 14 and Figure 15 Taking the transition of a wheeled quadruped robot from a stable two-wheeled state to a four-legged standing state on a pile surface (i.e., a supporting surface) as an example, the control method of the robot provided in this application embodiment will be described, and its specific content can be as follows:

[0137] In this embodiment, the process of switching from a two-wheel stable state to a four-legged standing state can be divided into two stages: the transition from a two-wheel stable state to a foot-wheel combined standing state (i.e., a foot-wheel combined prone state) and the transition from a foot-wheel combined standing state to a four-legged standing state (for the rear wheel, it is the transition from wheel support to foot support).

[0138] 1. Robot 1401 maintains a stable state on the pile surface by using the drive wheels on its two rear mechanical legs.

[0139] 2. Robot 1401 controls the drive wheels on its two rear mechanical legs to move backward, causing the torso to tilt forward.

[0140] 3. Robot 1401 controls the drive wheels on the two rear mechanical legs to stop rotating, and the torso begins to fall under the action of gravity. At the same time, robot 1401 gradually rotates each joint motor (such as the torso joint motor, hip joint motor and knee joint motor) to the desired landing angle required for landing, so that when the feet of the two mechanical legs land on the pile surface, the posture of robot 1401 meets the desired landing posture.

[0141] 4. Robot 1401 lands on the pile surface with its feet on its two front mechanical legs, while the drive wheels on its two rear mechanical legs remain on the original pile surface. The tilt angle corresponding to the torso is the expected tilt angle for landing, i.e., the preset value.

[0142] 5. Robot 1401 keeps its footing fixed and adjusts its torso to a horizontal position.

[0143] The above describes the transition from a stable two-wheeled stance to a combined foot-wheeled stance. The following describes the transition from a combined foot-wheeled stance to a quadrupedal stance.

[0144] 6. Robot 1401 switches the two rear mechanical legs from drive wheel motion mode to foot motion mode by adjusting the linear actuators on the two rear mechanical legs.

[0145] 7. Robot 1401 controls its four robotic legs to move its torso forward, gradually shifting its center of gravity towards the forward mechanical legs. When Robot 1401 has a target angular velocity, the drive wheels on the two rear mechanical legs detach from the pile surface under the inertial drive corresponding to the target angular velocity.

[0146] 8. Robot 1401 controls the two front mechanical legs to swing the torso backward, while adjusting the four mechanical legs to the angle corresponding to the desired posture at the end of the swing, so that the feet corresponding to the two rear mechanical legs can fall back onto the original pile surface.

[0147] 9. After the feet of the two hind mechanical legs touch the original pile surface, the robot 1401 makes a slight adjustment to its posture so that each joint returns to the angle required for stable four-legged standing.

[0148] 10. Robot 1401 levels its torso and enters a quadrupedal standing position.

[0149] In summary, the technical solution provided in this application allows the wheel-wheel quadruped robot to autonomously switch between a wheeled standing state and a legged crouching state even in situations with limited footing space (such as plum blossom stakes), thereby improving the applicability of the wheel-wheel quadruped robot.

[0150] Please refer to Figure 16 The diagram shows a flowchart of a robot control method provided in another embodiment of this application. The execution subject of each step of the method can be a robot (such as the above-mentioned wheeled quadruped robot). The method can include the following steps (1601-1603).

[0151] Step 1601: Adjust the tilt angle between the torso and the support surface to switch from a wheeled standing state to a foot-wheeled compound prone state. In the wheeled standing state, the robot maintains its standing position solely through the drive wheels corresponding to the first mechanical leg group. In the foot-wheeled compound prone state, the robot maintains its prone position through the drive wheels corresponding to the first mechanical leg group and the feet corresponding to the second mechanical leg group.

[0152] Step 1601 is the same as step 401 described above, and will not be repeated here.

[0153] Step 1602: Control the torso to tilt towards the first mechanical leg in the first mechanical leg group. When the second mechanical leg in the first mechanical leg group detaches from the support surface under the action of gravity, switch the second mechanical leg from the drive wheel ground support state to the foot ground support state.

[0154] The first and second mechanical legs can refer to the mechanical legs required for the robot to maintain its wheeled standing position.

[0155] In one example, the robot can adjust the angle of the joint motor corresponding to the first mechanical leg and the angle of the shutdown motor corresponding to at least two mechanical legs in the second mechanical leg group to move the projection of the center of gravity to a first area, which may refer to the area enclosed by the landing point of the first mechanical leg and the landing points of at least two mechanical legs in the second mechanical leg group.

[0156] When the robot's center of gravity projection moves into the first area, the robot tilts its torso towards the first mechanical leg under the influence of gravity, causing the second mechanical leg in the first mechanical leg assembly to begin to detach from the support surface. Optionally, during the process of the robot's center of gravity projection moving into the first area, the robot can also control the angle of the second mechanical leg's shutdown motor to cause the second mechanical leg to detach from the support surface.

[0157] When the second mechanical leg detaches from the support surface, the robot controls the second mechanical leg to switch from drive wheel movement mode to foot movement mode by adjusting the corresponding linear actuator.

[0158] When the robot's posture meets the first desired posture, the robot moves its torso towards the second mechanical leg by adjusting the angle of the shutdown motor corresponding to the first mechanical leg and the angle of the shutdown motor corresponding to at least two mechanical legs in the second mechanical leg group, so that the torso returns to the initial position corresponding to the foot-wheel compound prone position. The first desired posture can be set and adjusted based on empirical values.

[0159] During the process of restoring the torso to its initial position, the robot can adjust the angles of the motors at the joints of its second robotic leg to switch it from a drive wheel-based support state to a foot-based support state. This allows the robot to further stabilize its posture and enter a stable prone position.

[0160] For example, refer to Figure 17 and Figure 18 Taking the transition of a quadruped robot from a stable two-wheeled state to a quadrupedal standing state as an example, the control method of the robot provided in this application embodiment will be described, and its specific content is as follows:

[0161] 1. Robot 1701 enters the foot-wheel composite standing state.

[0162] 2. Robot 1701 adjusts the angle of the joint motor corresponding to the left rear mechanical leg and the angle of the shutdown motor corresponding to the two front mechanical legs, so that the projection of the center of gravity moves to the triangular area formed by the landing point corresponding to the left rear mechanical leg and the landing point corresponding to the two front mechanical legs, while controlling the right rear mechanical leg to lift off the ground.

[0163] 3. Robot 1701 controls the right rear mechanical leg to switch from drive wheel movement mode to foot movement mode by adjusting the linear actuator corresponding to the right rear mechanical leg.

[0164] 4. Robot 1701 adjusts the angle of the joint motor of the right rear mechanical leg to switch the right rear mechanical leg to a posture that allows it to stand upright.

[0165] 5. Robot 1701 adjusts the angle of the joint motor corresponding to the left rear mechanical leg and the angle of the shutdown motor corresponding to the two front mechanical legs to restore the torso to the initial position and switch the right rear mechanical leg from the drive wheel ground support state to the foot ground support state.

[0166] Step 1603: Control the torso to tilt towards the second mechanical leg. When the first mechanical leg detaches from the support surface under the action of gravity, switch the first mechanical leg from the drive wheel ground support state to the foot ground support state.

[0167] In one example, the robot can adjust the angle of the joint motor corresponding to the second mechanical leg and the angle of the shutdown motor corresponding to at least two mechanical legs in the second mechanical leg group to move the projection of the center of gravity to a second region. The second region can refer to the area enclosed by the landing point of the second mechanical leg and the landing points of at least two mechanical legs in the second mechanical leg group.

[0168] As the robot's center of gravity projection moves into the second region, the robot, under the influence of gravity, tilts its torso towards the second mechanical leg, causing the first mechanical leg to begin detaching from the support surface. Optionally, during the process of the robot's center of gravity projection moving into the second region, the robot can also control the angle of the first mechanical leg's shutdown motor to cause the first mechanical leg to detach from the support surface.

[0169] When the first mechanical leg detaches from the support surface, the robot controls the first mechanical leg to switch from drive wheel movement mode to foot movement mode by adjusting the linear actuator corresponding to the first mechanical leg.

[0170] When the robot's posture meets the second desired posture, the robot moves its torso towards the first mechanical leg by adjusting the angle of the shutdown motor corresponding to the second mechanical leg, as well as the angle of the shutdown motor corresponding to at least two mechanical legs in the second mechanical leg group, so that the torso returns to the initial position corresponding to the foot-wheel compound prone position. The second desired posture can be set and adjusted based on empirical values.

[0171] During the process of restoring the torso to its initial position, the robot can adjust the angles of the joint motors corresponding to the first mechanical leg, switching the first mechanical leg from a drive wheel-based support state to a foot-based support state. The robot can then further stabilize its body to enter a quadrupedal standing state.

[0172] For example, refer to Figure 17 and Figure 18 This process may specifically include the following:

[0173] 6. Robot 1701 adjusts the angle of the joint motor corresponding to the right rear mechanical leg and the angle of the shutdown motor corresponding to the two front mechanical legs, so that the projection of the center of gravity moves to the triangular area formed by the landing point corresponding to the right rear mechanical leg and the landing point corresponding to the two front mechanical legs, while controlling the left rear mechanical leg to lift off the ground.

[0174] 7. Robot 1701 controls the left rear mechanical leg to switch from drive wheel movement mode to foot movement mode by adjusting the linear actuator corresponding to the left rear mechanical leg.

[0175] 8. Robot 1701 adjusts the angle of the joint motor of the left rear mechanical leg to switch the left rear mechanical leg to a posture that allows it to stand upright.

[0176] 9. Robot 1701 adjusts the angle of the joint motor corresponding to the right rear mechanical leg and the angle of the shutdown motor corresponding to the two front mechanical legs to restore the torso to the initial position and switch the left rear mechanical leg from the drive wheel ground support state to the foot ground support state.

[0177] 10. Robot 1701 further stabilized its body to stably enter a quadrupedal standing position.

[0178] In summary, the technical solution provided in this application achieves the switching from a wheeled standing state to a foot-wheeled compound prone state by adjusting the tilt angle between the torso and the ground. Then, the mechanical legs in the first mechanical leg group are switched from a wheeled ground support state to a foot-ground support state, thereby realizing the switching from a wheeled standing state to a foot-ground prone state. This increases the robot's self-switching state method and improves the robot's self-switching state flexibility.

[0179] In addition, by adopting the technical solution provided in the embodiments of this application, the robot can switch from a wheeled standing state to a footed lying state in some special scenarios (such as a narrow footing area), which solves the problem that the robot cannot switch from a wheeled standing state to a footed lying state due to special scenarios. This expands the applicable scenarios of the robot and improves the robot's ability to switch states automatically, thereby improving the applicability of the robot.

[0180] In addition, for the wheel-wheel hybrid quadruped robot, the technical solution provided in the embodiments of this application enables the wheel-wheel hybrid quadruped robot to autonomously switch between wheeled standing state and legged lying state even in situations where the footing range is narrow (such as plum blossom stakes), thereby improving the applicability of the wheel-wheel hybrid quadruped robot.

[0181] In an exemplary embodiment, for the initial posture, desired landing posture, desired swing initiation posture, and desired swing termination posture described above, since the lengths of each structure of the robot and the distances between the landing points of each mechanical leg are fixed, and the desired tilt angle of the torso and the angle between the mechanical leg and the supporting surface can be set, the desired angles for each of the above postures can be calculated through geometric relationships. The following will use the method for obtaining the desired angle under the desired landing posture as an example for explanation. The methods for obtaining the desired angles under the initial posture, desired swing initiation posture, and desired swing termination posture can refer to the method for obtaining the desired angle under the desired landing posture, and will not be repeated here.

[0182] In the foot-wheel compound prone position, a coordinate system is constructed with the landing point of the first mechanical leg in the first mechanical leg assembly as the origin. For example, refer to... Figure 19 Robot 1901 (such as a quadruped robot with a combined foot and wheel) is in a combined foot and wheel lying position. A coordinate system is constructed with the landing point corresponding to the right rear mechanical leg of Robot 1901 as the origin. The horizontal axis of the coordinate system is parallel to the pile surface.

[0183] 1. Based on the angle between the torso and the horizontal axis, and the angle between the first mechanical leg and the horizontal axis, determine the expected landing angle corresponding to the hip joint motor of the first mechanical leg.

[0184] For example, refer to Figure 19 If the right rear mechanical leg of robot 1901 is perpendicular to the pile surface, then the angle between the right rear mechanical leg and the horizontal axis x is 90 degrees. If the angle between the torso of robot 1901 and the horizontal axis x is θ, then the expected landing angle corresponding to the hip joint motor of the right rear mechanical leg can be expressed as follows:

[0185] q h1 =θ+90.

[0186] 2. Based on the length of the torso, the radius of the drive wheel of the first mechanical leg, and the length from the drive wheel of the first mechanical leg to the torso, determine the first expected landing coordinates of the connection between the second mechanical leg and the torso.

[0187] For example, refer to Figure 19 The torso length of robot 1901 is l body The radius of the drive wheel is r wheel The thigh length of the right hind mechanical leg is l thigh Then, the expected first landing coordinates of the connection point between the right front mechanical leg and the torso can be represented as follows:

[0188] p fh =(x fh y fh )=(l body cos(θ), r wheel +l thigh +l body sin(θ));

[0189] Where, x fh Let y be the x-coordinate of the expected first landing coordinate. fh The ordinate is the expected coordinate of the first landing point.

[0190] 3. Based on the distance between the landing points of the first and second mechanical legs, determine the second expected landing coordinates corresponding to the landing point of the second mechanical leg.

[0191] For example, refer to Figure 19 The coordinates of the landing point of the right rear mechanical leg are (0, 0), and the distance between the landing point of the right rear mechanical leg and the landing point of the right front mechanical leg is l. gap Then, the second expected landing coordinates corresponding to the landing point of the right front mechanical leg can be represented as follows:

[0192] p ff =(x ff y ff )=(l gap ,0);

[0193] Where, x fh Let y be the x-coordinate of the expected second landing coordinates. fh The vertical coordinate is the expected coordinate of the second landing.

[0194] 4. Based on the first and second expected landing coordinates, obtain the landing point of the second mechanical leg and the expected landing distance between the hip joint motor of the second mechanical leg.

[0195] For example, refer to Figure 19 The expected landing distance between the landing point of the right front robotic leg and the hip joint motor of the right front robotic leg can be expressed as follows:

[0196] l fh，ff =((x) ff -x fh ) 2 +(y ff -y fh ) 2 ) 1 / 2 .

[0197] 5. Based on the expected landing distance, the distance between the foot of the second mechanical leg and the knee joint of the second mechanical leg, and the distance between the knee joint of the second mechanical leg and the torso, determine the second expected landing angle corresponding to the knee joint motor of the second mechanical leg.

[0198] For example, refer to Figure 19 According to the l corresponding to the right front mechanical leg thigh l thigh (calf length) and l fh，ff Using the law of cosines, the second expected landing angle corresponding to the knee joint motor of the right front mechanical leg can be obtained:

[0199] q f2 =arccos((l thigh 2 +l shank 2 -l fh，ff 2 ) / (2l thigh l shank )).

[0200] 6. Based on the angle between the torso and the horizontal axis, the expected landing distance, the first expected landing coordinate, the distance between the foot of the second mechanical leg and the knee joint of the second mechanical leg, and the distance between the knee joint of the second mechanical leg and the torso, determine the third expected landing angle corresponding to the hip joint motor of the second mechanical leg.

[0201] For example, refer to Figure 19 The third expected landing angle corresponding to the hip joint motor of the right front mechanical leg can be expressed as follows:

[0202] q f1 =θ+arcsin(y fh / l fh , ff )+arccos((l thigh 2 +l fh，ff 2 -l shank 2 ) / (2l thigh l fh，ff )).

[0203] After obtaining these desired angles, the corresponding desired angular velocities are obtained through differential calculation. The robot then sends these desired angle and desired angular velocity commands to the corresponding joint motors, enabling the joint motors to follow the desired angles and desired angular velocities, thereby achieving the switching from a wheeled standing state to a legged prone state.

[0204] In one exemplary embodiment, the robot includes a position loop controller, a velocity loop controller, and a current loop controller, and the robot's control strategy is as follows:

[0205] 1. Obtain the desired angle, desired angular velocity, and desired torque for each joint motor of the robot.

[0206] Optionally, the desired angle, desired angular velocity, and desired torque corresponding to each joint motor can refer to the desired angle, desired angular velocity, and desired torque corresponding to each joint motor under the aforementioned initial posture, landing desired posture, swing start desired posture, and swing end desired posture, etc.

[0207] 2. Using the position loop controller, based on the desired angle corresponding to each joint motor and the angle feedback information corresponding to each joint motor, the desired angle corresponding to each joint motor is followed.

[0208] For example, refer to Figure 20 The robot can follow the desired angle corresponding to each joint motor by controlling the machine through the position loop based on the desired angle corresponding to each joint motor and the position feedback (i.e., angle feedback information) corresponding to each joint motor.

[0209] 3. Using a speed loop controller, based on the desired angular velocity of each joint motor and the feedback information of the angular velocity of each joint motor, the desired angular velocity of each joint motor is followed.

[0210] For example, refer to Figure 20 The robot can follow the desired angular velocity of each joint motor by using a velocity loop controller, based on the desired angular velocity of each joint motor and the velocity feedback (i.e., angular velocity feedback information) of each joint motor.

[0211] 4. Using a current loop controller, based on the desired torque corresponding to each joint motor and the torque feedback information corresponding to each joint motor, the desired torque corresponding to each joint motor is followed.

[0212] For example, refer to Figure 20The robot can follow the expected torque of each joint motor by using a current loop controller, based on the expected torque (i.e., feedforward torque) and the torque feedback information (i.e., current feedback) of each joint motor.

[0213] Optionally, the desired angle, desired angular velocity, and desired torque are processed by the position loop controller 2001, the instantaneous disconnection loop controller 2002, and the current loop controller 2003 to obtain the joint motor drive commands corresponding to each joint motor. Each joint motor is driven according to its respective joint motor drive command to realize the movement of the robot.

[0214] In summary, the technical solution provided in this application, by adjusting the tilt angle between the torso and the support surface, achieves the switching from a wheeled standing state to a foot-wheeled compound prone state. Then, the mechanical legs in the first mechanical leg group are switched from a wheeled ground support state to a foot-ground support state, thereby realizing the switching from a wheeled standing state to a foot-ground prone state. This increases the robot's self-switching state method and improves the robot's self-switching state flexibility.

[0215] In addition, by adopting the technical solution provided in the embodiments of this application, the robot can switch from a wheeled standing state to a footed lying state in some special scenarios (such as a narrow footing area), which solves the problem that the robot cannot switch from a wheeled standing state to a footed lying state due to special scenarios. This expands the applicable scenarios of the robot and improves the robot's ability to switch states automatically, thereby improving the applicability of the robot.

[0216] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0217] refer to Figure 21 This diagram illustrates a block diagram of a robot control device according to an embodiment of this application. The device has the function of implementing the robot control method described above; this function can be implemented in hardware or by hardware executing corresponding software. The device can be a robot as described above (such as a wheeled quadruped robot) or can be housed within a robot. Figure 21 As shown, the device 2100 includes: a foot wheel ground-suspension switching module 2101, a drive wheel off-ground module 2102, and a foot ground-suspension switching module 2103.

[0218] The wheel-mounted prone switching module 2101 is used to adjust the tilt angle between the torso and the support surface, switching from a wheeled standing state to a wheeled compound prone state. In the wheeled standing state, the robot maintains its standing position solely through the drive wheels corresponding to the first mechanical leg group. In the wheeled compound prone state, the robot maintains its prone position through the drive wheels corresponding to the first mechanical leg group and the feet corresponding to the second mechanical leg group.

[0219] The drive wheel off-ground module 2102 is used to swing the torso in a first direction with the foot corresponding to the second mechanical leg group as the center, and control the drive wheel corresponding to the first mechanical leg group to detach from the support surface.

[0220] The foot-prone switching module 2103 is used to swing the torso in a second direction with the foot corresponding to the second mechanical leg group as the center, and control the foot corresponding to the first mechanical leg group to land and enter a foot-prone state; wherein, in the foot-prone state, the robot only maintains a prone position through its feet; wherein, the first direction is opposite to the second direction.

[0221] In one exemplary embodiment, the drive wheel ground clearance module 2102 is used for:

[0222] Maintaining the robot's footing, the robot moves its torso in the first direction, switching from the foot-wheel compound prone state to the swing preparation state; wherein, in the swing preparation state, the robot has a target angular velocity that causes the drive wheel corresponding to the first mechanical leg assembly to detach from the support surface;

[0223] Driven by the inertia corresponding to the target angular velocity, the torso is controlled to swing in the first direction with the foot corresponding to the second mechanical leg group as the center, and the drive wheel corresponding to the first mechanical leg group disengages from the support surface.

[0224] In one exemplary embodiment, the drive wheel ground clearance module 2102 is further configured to:

[0225] Obtain the expected swing posture corresponding to the robot. The expected swing posture includes the expected swing angle between the mechanical leg in the first mechanical leg group and the support surface, the expected swing angle of each joint motor of the robot in the swing preparation state, and the expected swing tilt angle between the torso and the support surface.

[0226] Based on the posture corresponding to the foot wheel compound ground state and the desired swing posture, the corresponding swing preparation posture sequence of the robot is determined; wherein, the swing preparation posture sequence includes the desired posture of the robot at each moment in the swing preparation stage, and the desired angle between the mechanical leg in the first mechanical leg group and the support surface at each moment in the swing preparation stage.

[0227] According to the swing preparation posture sequence, maintain the landing point of the robot, and adjust the angle between the mechanical leg in the first mechanical leg group and the support surface and the angle of each joint motor to control the torso to move in the first direction;

[0228] When the robot's angular velocity meets the target angular velocity, it enters the swing preparation state.

[0229] In one exemplary embodiment, the foot-prone switching module 2103 is used for:

[0230] Adjust the linear actuator corresponding to the first mechanical leg assembly to control the first mechanical leg assembly to switch from drive wheel movement mode to foot movement mode;

[0231] With the foot corresponding to the second mechanical leg group as the center, adjust the angle of the joint motor corresponding to the second mechanical leg group to swing the torso in the second direction;

[0232] Based on the desired swing termination posture of the robot, the angles of the motors of each joint of the robot are adjusted to control the foot of the first mechanical leg group to land; wherein, the desired swing termination posture includes the desired termination joint angle of the foot of the first mechanical leg group.

[0233] Enter the foot-prone position.

[0234] In one exemplary embodiment, the wheel-mounted prone switching module 2101 is used for:

[0235] Obtain the initial posture corresponding to the wheeled standing state, the initial posture including the initial angles of each joint motor of the robot in the wheeled standing state;

[0236] Obtain the expected landing posture corresponding to the foot wheel compound prone state. The expected landing posture includes the expected landing angle of each joint motor in the foot wheel compound prone state, and the expected landing tilt angle between the torso and the support surface.

[0237] Based on the initial posture and the expected landing posture, a landing planning posture sequence corresponding to the robot is obtained, and the landing planning posture sequence includes the expected posture of the robot at each moment in the landing phase.

[0238] According to the landing planning posture sequence, adjust the tilt angle between the torso and the support surface, and control the robot's center of gravity to move in the first direction;

[0239] When the robot's center of gravity satisfies the torso's descent, the angles of each joint motor are adjusted according to the landing posture sequence, and the robot enters the foot-wheel compound prone state upon landing.

[0240] In an exemplary embodiment, the foot-wheel-crawling switching module 2101 is further configured to control the drive wheel corresponding to the first mechanical leg group to move in the second direction according to the landing planning posture sequence, the tilt angle between the torso and the support surface increases under the action of inertia, and the center of gravity of the robot moves in the first direction.

[0241] In an exemplary embodiment, each of the robot's mechanical legs is equipped with a hip joint motor and a knee joint motor. In the foot-wheel compound prone state, a coordinate system is constructed with the landing point of the first mechanical leg in the first mechanical leg group as the origin. The method for obtaining the expected landing angles corresponding to the first mechanical leg and the second mechanical leg on the same side as the first mechanical leg in the second mechanical leg group is as follows:

[0242] Based on the angle between the torso and the horizontal axis, and the angle between the first mechanical leg and the horizontal axis, the expected landing angle corresponding to the hip joint motor of the first mechanical leg is determined.

[0243] Based on the length of the torso, the radius of the drive wheel of the first mechanical leg, and the distance from the drive wheel of the first mechanical leg to the torso, the first expected landing coordinates of the connection between the second mechanical leg and the torso are determined.

[0244] Based on the distance between the landing point of the first mechanical leg and the landing point of the second mechanical leg, the second expected landing coordinates corresponding to the landing point of the second mechanical leg are determined.

[0245] Based on the first expected landing coordinates and the second expected landing coordinates, the expected landing distance between the landing point of the second mechanical leg and the hip joint motor of the second mechanical leg is obtained;

[0246] Based on the expected landing distance, the distance between the foot of the second mechanical leg and the knee joint of the second mechanical leg, and the distance between the knee joint of the second mechanical leg and the torso, the second expected landing angle corresponding to the knee joint motor of the second mechanical leg is determined.

[0247] Based on the angle between the torso and the horizontal axis, the expected landing distance, the first expected landing coordinate, the distance between the foot of the second mechanical leg and the knee joint of the second mechanical leg, and the distance between the knee joint of the second mechanical leg and the torso, the third expected landing angle corresponding to the hip joint motor of the second mechanical leg is determined.

[0248] In an exemplary embodiment, the robot includes a position loop controller, a velocity loop controller, and a current loop controller, and the control strategy of the robot is as follows:

[0249] Obtain the desired angle, desired angular velocity, and desired torque for each joint motor of the robot;

[0250] The position loop controller tracks the desired angles of each joint motor based on the desired angles corresponding to each joint motor and the angle feedback information corresponding to each joint motor.

[0251] The speed loop controller tracks the desired angular velocity of each joint motor based on the desired angular velocity of each joint motor and the angular velocity feedback information of each joint motor.

[0252] The current loop controller tracks the desired torque of each joint motor based on the desired torque and torque feedback information of each joint motor.

[0253] In one exemplary embodiment, the landing range of the mechanical leg in the first mechanical leg group is less than a threshold range.

[0254] In one exemplary embodiment, the robot is a quadruped robot with a wheel and a foot, and the robot switches from a two-wheeled balanced state to a quadrupedal lying state.

[0255] In summary, the technical solution provided in this application, by adjusting the tilt angle between the torso and the support surface, achieves the switching from a wheeled standing state to a foot-wheeled compound prone state. Then, based on the second mechanical leg assembly, the torso is swung, controlling the drive wheel corresponding to the first mechanical leg assembly to detach from the support surface. During the swinging process, the drive wheel's ground support is switched to the foot's ground support, thereby achieving the switching from a wheeled standing state to a foot-wheeled prone state. This increases the robot's self-switching state method and improves the robot's self-switching state flexibility.

[0256] In addition, by adopting the technical solution provided in the embodiments of this application, the robot can switch from a wheeled standing state to a footed lying state in some special scenarios (such as a narrow footing area), which solves the problem that the robot cannot switch from a wheeled standing state to a footed lying state due to special scenarios. This expands the applicable scenarios of the robot and improves the robot's ability to switch states automatically, thereby improving the applicability of the robot.

[0257] In addition, for the wheel-wheel hybrid quadruped robot, the technical solution provided in the embodiments of this application enables the wheel-wheel hybrid quadruped robot to autonomously switch between wheeled standing state and legged lying state even in situations where the footing range is narrow (such as plum blossom stakes), thereby improving the applicability of the wheel-wheel hybrid quadruped robot.

[0258] refer to Figure 22 This diagram illustrates a block diagram of a robot control device according to an embodiment of this application. The device has the function of implementing the robot control method described above; this function can be implemented in hardware or by hardware executing corresponding software. The device can be a robot as described above (such as a wheeled quadruped robot), or it can be integrated into a robot. Figure 22 As shown, the device 2200 includes: a foot wheel-based prone switching module 2201, a first foot switching module 2202, and a second foot switching module 2203.

[0259] The wheel-mounted prone switching module 2201 is used to adjust the tilt angle between the torso and the support surface, switching from a wheeled standing state to a wheeled compound prone state. In the wheeled standing state, the robot maintains its standing position solely through the drive wheels corresponding to the first mechanical leg group. In the wheeled compound prone state, the robot maintains its prone position through the drive wheels corresponding to the first mechanical leg group and the feet corresponding to the second mechanical leg group.

[0260] The first foot switching module 2202 is used to control the torso to tilt towards the first mechanical leg in the first mechanical leg group, and to switch the second mechanical leg from the drive wheel ground support state to the foot ground support state when the second mechanical leg in the first mechanical leg group leaves the support surface under the action of gravity.

[0261] The second foot switching module 2203 is used to control the torso to tilt towards the second mechanical leg, and to switch the first mechanical leg from the drive wheel ground support state to the foot ground support state when the first mechanical leg leaves the support surface under the action of gravity.

[0262] In summary, the technical solution provided in this application, by adjusting the tilt angle between the torso and the support surface, achieves the switching from a wheeled standing state to a foot-wheeled compound prone state. Then, the mechanical legs in the first mechanical leg group are switched from a wheeled ground support state to a foot-ground support state, thereby realizing the switching from a wheeled standing state to a foot-ground prone state. This increases the robot's self-switching state method and improves the robot's self-switching state flexibility.

[0263] In addition, by adopting the technical solution provided in the embodiments of this application, the robot can switch from a wheeled standing state to a footed lying state in some special scenarios (such as a narrow footing area), which solves the problem that the robot cannot switch from a wheeled standing state to a footed lying state due to special scenarios. This expands the applicable scenarios of the robot and improves the robot's ability to switch states automatically, thereby improving the applicability of the robot.

[0264] In addition, for the wheel-wheel hybrid quadruped robot, the technical solution provided in the embodiments of this application enables the wheel-wheel hybrid quadruped robot to autonomously switch between wheeled standing state and legged lying state even in situations where the footing range is narrow (such as plum blossom stakes), thereby improving the applicability of the wheel-wheel hybrid quadruped robot.

[0265] It should be noted that the apparatus provided in the above embodiments is only illustrated by the division of the above functional modules when implementing its functions. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0266] Please refer to Figure 23 This diagram illustrates a simplified structural block diagram of a robot provided in one embodiment of this application. The robot may be a quadruped robot, a hexapod robot, etc., and this embodiment does not limit the scope of the application.

[0267] Optionally, such as Figure 23 As shown, the robot includes a processor 2301 and a memory 2302. The processor 2301 includes, but is not limited to, any of the following: CPU (Central Processing Unit), GPU (Graphics Processing Unit), and FPGA (Field Programmable Gate Array). The memory 2302 may include storage devices such as RAM (Random-Access Memory) and ROM (Read-Only Memory). The processor 2301 and the memory 2302 can be connected via a system bus.

[0268] In an exemplary embodiment, the memory 2302 stores a computer program, which is loaded and executed by the processor 2301 to implement the robot control method described above.

[0269] In an exemplary embodiment, a computer-readable storage medium is also provided, wherein a computer program is stored therein, which, when executed by the robot's processor, implements the robot control method described above.

[0270] Optionally, the computer-readable storage medium may include: ROM (Read-Only Memory), RAM (Random-Access Memory), SSD (Solid State Drives), or optical disc, etc. The random access memory may include ReRAM (Resistance Random Access Memory) and DRAM (Dynamic Random Access Memory).

[0271] In an exemplary embodiment, a computer program product or computer program is also provided, the computer program product or computer program including computer instructions stored in a computer-readable storage medium. A robot's processor reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, causing the robot to perform the aforementioned robot control method.

[0272] It should be noted that all information (including but not limited to object device information, object personal information, etc.), data (including but not limited to data used for analysis, stored data, displayed data, etc.), and signals involved in this application have been authorized by the object or fully authorized by all parties, and the collection, use, and processing of related data must comply with the relevant laws, regulations, and standards of the relevant countries and regions. For example, the robot architecture and robot control methods involved in this application were obtained with full authorization.

[0273] It should be understood that "multiple" as used herein refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. Furthermore, the step numbers described herein are merely illustrative of one possible execution order. In some other embodiments, the steps may not be executed in numerical order, such as two steps with different numbers being executed simultaneously, or two steps with different numbers being executed in the reverse order of the illustration. This application does not limit this.

[0274] The above description is merely an exemplary embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for controlling a robot, characterized in that, The robot includes a torso, and a first mechanical leg assembly and a second mechanical leg assembly connected to the torso. The first mechanical leg assembly includes at least two wheel-driven mechanical legs, which move via drive wheels or feet. The method includes: Adjust the tilt angle between the torso and the support surface to switch from a wheeled standing state to a foot-wheeled compound prone state; wherein, in the wheeled standing state, the robot maintains its standing position solely through the drive wheels corresponding to the first mechanical leg group; in the foot-wheeled compound prone state, the robot maintains its prone position through the drive wheels corresponding to the first mechanical leg group and the feet corresponding to the second mechanical leg group; Maintaining the robot's footing, the robot moves its torso in the first direction, switching from the foot-wheel compound prone state to the swing preparation state; wherein, in the swing preparation state, the robot has a target angular velocity that causes the drive wheel corresponding to the first mechanical leg group to detach from the support surface; Driven by the inertia corresponding to the target angular velocity, the torso is controlled to swing in the first direction with the foot corresponding to the second mechanical leg group as the center, and the drive wheel corresponding to the first mechanical leg group disengages from the support surface; The torso swings in a second direction with the foot corresponding to the second mechanical leg group as the center, controlling the foot corresponding to the first mechanical leg group to land and enter a footed prone state; wherein, in the footed prone state, the robot maintains its prone position only through its feet; Wherein, the first direction is opposite to the second direction.

2. The method according to claim 1, characterized in that, Maintaining the robot's foothold and moving its torso in the first direction to switch from the wheeled compound prone state to the swing preparation state includes: Obtain the expected swing posture corresponding to the robot. The expected swing posture includes the expected swing angle between the mechanical leg in the first mechanical leg group and the support surface, the expected swing angle of each joint motor of the robot in the swing preparation state, and the expected swing tilt angle between the torso and the support surface. Based on the posture corresponding to the foot wheel compound ground state and the desired swing posture, the corresponding swing preparation posture sequence of the robot is determined; wherein, the swing preparation posture sequence includes the desired posture of the robot at each moment in the swing preparation stage, and the desired angle between the mechanical leg in the first mechanical leg group and the support surface at each moment in the swing preparation stage. According to the swing preparation posture sequence, maintain the landing point of the robot, and adjust the angle between the mechanical leg in the first mechanical leg group and the support surface and the angle of each joint motor to control the torso to move in the first direction; When the robot's angular velocity meets the target angular velocity, it enters the swing preparation state.

3. The method according to claim 1, characterized in that, The step of swinging the torso in a second direction with the foot corresponding to the second mechanical leg group as the center, and controlling the foot corresponding to the first mechanical leg group to land on the ground, and entering a footed prone state, includes: Adjust the linear actuator corresponding to the first mechanical leg assembly to control the first mechanical leg assembly to switch from drive wheel movement mode to foot movement mode; With the foot corresponding to the second mechanical leg group as the center, adjust the angle of the joint motor corresponding to the second mechanical leg group to swing the torso in the second direction; Based on the desired swing termination posture of the robot, the angles of the motors of each joint of the robot are adjusted to control the foot of the first mechanical leg group to land; wherein, the desired swing termination posture includes the desired termination joint angle of the foot of the first mechanical leg group. Enter the foot-prone position.

4. The method according to claim 1, characterized in that, The adjustment of the tilt angle between the torso and the support surface, switching from a wheeled standing state to a wheeled compound prone state, includes: Obtain the initial posture corresponding to the wheeled standing state, the initial posture including the initial angles of each joint motor of the robot in the wheeled standing state; Obtain the expected landing posture corresponding to the foot wheel compound prone state. The expected landing posture includes the expected landing angle of each joint motor in the foot wheel compound prone state, and the expected landing tilt angle between the torso and the support surface. Based on the initial posture and the expected landing posture, a landing planning posture sequence corresponding to the robot is obtained, and the landing planning posture sequence includes the expected posture of the robot at each moment in the landing phase. According to the landing planning posture sequence, adjust the tilt angle between the torso and the support surface, and control the robot's center of gravity to move in the first direction; When the robot's center of gravity satisfies the torso's descent, the angles of each joint motor are adjusted according to the landing posture sequence, and the robot enters the foot-wheel compound prone state upon landing.

5. The method according to claim 4, characterized in that, The step of adjusting the tilt angle between the robot's torso and the support surface according to the landing planned posture sequence, and controlling the robot's center of gravity to move in the first direction, includes: According to the landing planning posture sequence, the drive wheel corresponding to the first mechanical leg group is controlled to move in the second direction, the tilt angle between the torso and the support surface increases under the action of inertia, and the center of gravity of the robot moves in the first direction.

6. The method according to any one of claims 1 to 5, characterized in that, Each of the robot's mechanical legs is equipped with a hip joint motor and a knee joint motor. In the foot-wheel compound prone state, a coordinate system is constructed with the landing point of the first mechanical leg in the first mechanical leg group as the origin. The method for obtaining the expected landing angles of the first mechanical leg and the second mechanical leg on the same side as the first mechanical leg in the second mechanical leg group is as follows: Based on the angle between the torso and the horizontal axis, and the angle between the first mechanical leg and the horizontal axis, the expected landing angle corresponding to the hip joint motor of the first mechanical leg is determined. Based on the length of the torso, the radius of the drive wheel of the first mechanical leg, and the distance from the drive wheel of the first mechanical leg to the torso, the first expected landing coordinates of the connection between the second mechanical leg and the torso are determined. Based on the distance between the landing point of the first mechanical leg and the landing point of the second mechanical leg, the second expected landing coordinates corresponding to the landing point of the second mechanical leg are determined. Based on the first expected landing coordinates and the second expected landing coordinates, the expected landing distance between the landing point of the second mechanical leg and the hip joint motor of the second mechanical leg is obtained; Based on the expected landing distance, the distance between the foot of the second mechanical leg and the knee joint of the second mechanical leg, and the distance between the knee joint of the second mechanical leg and the torso, the second expected landing angle corresponding to the knee joint motor of the second mechanical leg is determined. Based on the angle between the torso and the horizontal axis, the expected landing distance, the first expected landing coordinate, the distance between the foot of the second mechanical leg and the knee joint of the second mechanical leg, and the distance between the knee joint of the second mechanical leg and the torso, the third expected landing angle corresponding to the hip joint motor of the second mechanical leg is determined.

7. The method according to any one of claims 1 to 5, characterized in that, The robot includes a position loop controller, a velocity loop controller, and a current loop controller. The control strategy of the robot is as follows: Obtain the desired angle, desired angular velocity, and desired torque for each joint motor of the robot; The position loop controller tracks the desired angles of each joint motor based on the desired angles corresponding to each joint motor and the angle feedback information corresponding to each joint motor. The speed loop controller tracks the desired angular velocity of each joint motor based on the desired angular velocity of each joint motor and the angular velocity feedback information of each joint motor. The current loop controller tracks the desired torque of each joint motor based on the desired torque and torque feedback information of each joint motor.

8. The method according to any one of claims 1 to 5, characterized in that, The landing range of the mechanical leg in the first mechanical leg group is less than the threshold range.

9. The method according to any one of claims 1 to 5, characterized in that, The robot is a quadruped robot with a combination of wheels and legs, and it can switch from a two-wheeled balanced state to a four-legged prone state.

10. A method for controlling a robot, characterized in that, The robot includes a torso, and a first mechanical leg assembly and a second mechanical leg assembly connected to the torso. The first mechanical leg assembly includes at least two wheel-driven mechanical legs, which move via drive wheels or feet. The method includes: Adjust the tilt angle between the torso and the support surface to switch from a wheeled standing state to a foot-wheeled compound prone state; wherein, in the wheeled standing state, the robot maintains its standing position solely through the drive wheels corresponding to the first mechanical leg group; in the foot-wheeled compound prone state, the robot maintains its prone position through the drive wheels corresponding to the first mechanical leg group and the feet corresponding to the second mechanical leg group; Control the torso to tilt towards the first mechanical leg in the first mechanical leg group. When the second mechanical leg in the first mechanical leg group detaches from the support surface under the action of gravity, switch the second mechanical leg from the drive wheel ground support state to the foot ground support state. The torso is tilted toward the second mechanical leg. When the first mechanical leg detaches from the support surface under the action of gravity, the first mechanical leg is switched from the state where the drive wheel is on the ground to the state where the foot is on the ground.

11. A control device for a robot, characterized in that, The robot includes a torso, and a first mechanical leg assembly and a second mechanical leg assembly connected to the torso. The first mechanical leg assembly includes at least two wheel-driven mechanical legs, which move via drive wheels or feet. The device includes: The wheel-mounted prone switching module is used to adjust the tilt angle between the torso and the support surface, switching from a wheeled standing state to a wheeled compound prone state. In the wheeled standing state, the robot maintains its standing position solely through the drive wheels corresponding to the first mechanical leg group; in the wheeled compound prone state, the robot maintains its prone position through the drive wheels corresponding to the first mechanical leg group and the feet corresponding to the second mechanical leg group. The drive wheel lift-off module is used to maintain the robot's footing point, move the torso in a first direction, and switch from the foot-wheel compound prone state to a swing preparation state; wherein, in the swing preparation state, the robot has a target angular velocity that causes the drive wheel corresponding to the first mechanical leg group to lift off the support surface; driven by the inertia corresponding to the target angular velocity, the torso is controlled to swing in the first direction with the foot corresponding to the second mechanical leg group as the center, and the drive wheel corresponding to the first mechanical leg group lifts off the support surface; The foot-prone switching module is used to swing the torso in a second direction with the foot corresponding to the second mechanical leg group as the center, and control the foot corresponding to the first mechanical leg group to land and enter a foot-prone state; wherein, in the foot-prone state, the robot only maintains a prone position through its feet; wherein, the first direction is opposite to the second direction.

12. A control device for a robot, characterized in that, The robot includes a torso, and a first mechanical leg assembly and a second mechanical leg assembly connected to the torso. The first mechanical leg assembly includes at least two wheel-driven mechanical legs, which move via drive wheels or feet. The device includes: The wheel-mounted prone switching module is used to adjust the tilt angle between the torso and the support surface, switching from a wheeled standing state to a wheeled compound prone state. In the wheeled standing state, the robot maintains its standing position solely through the drive wheels corresponding to the first mechanical leg group; in the wheeled compound prone state, the robot maintains its prone position through the drive wheels corresponding to the first mechanical leg group and the feet corresponding to the second mechanical leg group. The first foot switching module is used to control the torso to tilt towards the first mechanical leg in the first mechanical leg group. When the second mechanical leg in the first mechanical leg group leaves the support surface under the action of gravity, the second mechanical leg is switched from the drive wheel ground support state to the foot ground support state. The second foot switching module is used to control the torso to tilt towards the second mechanical leg, and to switch the first mechanical leg from the drive wheel ground support state to the foot ground support state when the first mechanical leg detaches from the support surface under the action of gravity.

13. A robot, characterized in that, The robot includes a processor and a memory, the memory storing a computer program, which is loaded and executed by the processor to implement the robot control method as described in any one of claims 1 to 9, or to implement the robot control method as described in claim 10.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which is loaded and executed by a processor to implement the robot control method as described in any one of claims 1 to 9, or to implement the robot control method as described in claim 10.

15. A computer program product, characterized in that, The computer program product includes computer instructions stored in a computer-readable storage medium, and a processor reads and executes the computer instructions from the computer-readable storage medium to implement the robot control method as described in any one of claims 1 to 9, or to implement the robot control method as described in claim 10.

Images