Robot control method and device, behavior construction method and device and robot

Abstract

The invention relates to a robot control method and device, a behavior construction method and device and a robot. The control method comprises the steps that the actual contact states of a plurality of sole contact point positions of the robot are obtained; and according to the action sequence of the target behavior and the actual contact states of the plurality of sole contact points, controlling the robot to complete the target behavior through movement. The motion sequence comprises a plurality of motion stages, the motion stages comprise theoretical contact states of a plurality of sole contact points, and at least one sole contact point with different theoretical contact states exists between different motion stages, so that the motion sequence can describe target motion more accurately and closely; particularly, compared with the condition that the sole only comprises one contact point, the anthropomorphic level of the action sequence is improved; therefore, the motion of the robot is controlled based on the motion sequence and the actual contact state of the plurality of sole contact points, and the complexity, anthropomorphic level and stability of the motion of the robot can be improved.

Description

Technical Field

[0001] The present disclosure relates to the field of robot technology, and in particular to a robot control method and device, a behavior construction method and device, and a robot. Background Art

[0002] In recent years, the technology of robotics has been developing continuously, becoming more and more intelligent and automated, and the richness, stability and flexibility of actions have been improved to varying degrees. Robots can replace users in production and life to perform specific tasks, thus bringing convenience to users. In related technologies, the actions that robots can perform are relatively simple, that is, they are not yet able to perform more complex actions. Summary of the invention

[0003] In order to overcome the problems existing in the related art, the embodiments of the present disclosure provide a robot control method, device, robot and storage medium to solve the defects in the related art.

[0004] According to a first aspect of an embodiment of the present disclosure, there is provided a robot control method, the method comprising:

[0005] Acquire the actual contact status of multiple sole contact points of the robot, wherein the actual contact status includes touching the ground and not touching the ground;

[0006] According to the action sequence of the target behavior and the actual contact states of the multiple sole contact points, the robot is controlled to complete the target behavior through movement, wherein the action sequence includes multiple action stages, the action stages include theoretical contact states of the multiple sole contact points, and there is at least one sole contact point with a different theoretical contact state between different action stages.

[0007] In one embodiment of the present disclosure, the sole of the robot has a plurality of contact sensors, wherein the sole contact point corresponds to at least one contact sensor among the plurality of contact sensors;

[0008] The obtaining of the actual contact states of the multiple sole contact points of the robot includes:

[0009] The contact parameters collected by the multiple contact sensors are acquired, and the contact state of the sole contact point is correspondingly determined according to the contact parameters collected by at least one contact sensor corresponding to the sole contact point.

[0010] In one embodiment of the present disclosure, the method further includes:

[0011] Based on the changes of the theoretical contact states of the multiple sole contact points in the target behavior, an action sequence of the target behavior is constructed, wherein adjacent action stages in the action sequence are divided based on contact state switching events or contact state switching times.

[0012] In one embodiment of the present disclosure, the plurality of sole contact points include at least one point combination, wherein the point combination includes at least one sole contact point;

[0013] In each action phase of the action sequence, the theoretical contact states of multiple sole contact points in the same point combination are the same;

[0014] The obtaining of the actual contact states of the multiple sole contact points of the robot includes:

[0015] The actual contact state of any sole contact point in each point combination in the at least one point combination is obtained as the actual contact state of the corresponding point combination.

[0016] In one embodiment of the present disclosure, the plurality of sole contact points belong to the foot ends of the robot, and the number of sole contact points of each foot end is the same.

[0017] In one embodiment of the present disclosure, the robot is a bipedal robot; and / or the foot end has four sole contact points distributed at the upper left, lower left, upper right, and lower right.

[0018] In one embodiment of the present disclosure, the target behavior includes a jumping behavior, and the action sequence of the target behavior includes a full-foot contact phase, a forefoot contact phase, a flight phase, a rear-foot contact phase, and a full-foot contact phase arranged in sequence.

[0019] According to a second aspect of an embodiment of the present disclosure, a behavior construction method is provided, the method comprising:

[0020] Based on the changes in the theoretical contact states of multiple plantar contact points of the robot in the target behavior, an action sequence of the target behavior is constructed, wherein the action sequence includes multiple action stages, and the action stages include the theoretical contact states of the multiple plantar contact points. There is at least one plantar release point with a different theoretical contact state between different action stages, and adjacent action stages in the action sequence are divided based on contact state switching events or contact state switching times.

[0021] According to a third aspect of an embodiment of the present disclosure, there is provided a robot control device, the device comprising:

[0022] An acquisition module, used to acquire actual contact states of multiple sole contact points of the robot, wherein the actual contact states include ground contact and non-ground contact;

[0023] A control module is used to control the robot to complete the target behavior through movement according to the action sequence of the target behavior and the actual contact states of the multiple sole contact points, wherein the action sequence includes multiple action stages, and the action stages include theoretical contact states of the multiple sole contact points, and there is at least one sole contact point with a different theoretical contact state between different action stages.

[0024] In one embodiment of the present disclosure, the sole of the robot has a plurality of contact sensors, wherein the sole contact point corresponds to at least one contact sensor among the plurality of contact sensors;

[0025] The acquisition module is used for:

[0026] The contact parameters collected by the multiple contact sensors are acquired, and the contact state of the sole contact point is correspondingly determined according to the contact parameters collected by at least one contact sensor corresponding to the sole contact point.

[0027] In one embodiment of the present disclosure, the apparatus further comprises a building block for:

[0028] Based on the changes of the theoretical contact states of the multiple sole contact points in the target behavior, an action sequence of the target behavior is constructed, wherein adjacent action stages in the action sequence are divided based on contact state switching events or contact state switching times.

[0029] In one embodiment of the present disclosure, the plurality of sole contact points include at least one point combination, wherein the point combination includes at least one sole contact point;

[0030] In each action phase of the action sequence, the theoretical contact states of multiple sole contact points in the same point combination are the same;

[0031] The acquisition module is used for:

[0032] The actual contact state of any sole contact point in each point combination in the at least one point combination is obtained as the actual contact state of the corresponding point combination.

[0033] In one embodiment of the present disclosure, the plurality of sole contact points belong to the foot ends of the robot, and the number of sole contact points of each foot end is the same.

[0034] In one embodiment of the present disclosure, the robot is a bipedal robot; and / or the foot end has four sole contact points distributed at the upper left, lower left, upper right, and lower right.

[0035] In one embodiment of the present disclosure, the target behavior includes a jumping behavior, and the action sequence of the target behavior includes a full-foot contact phase, a forefoot contact phase, a flight phase, a rear-foot contact phase, and a full-foot contact phase arranged in sequence.

[0036] According to a fourth aspect of an embodiment of the present disclosure, a behavior construction device is provided, the device comprising:

[0037] A construction module is used to construct an action sequence of the target behavior based on changes in the theoretical contact states of multiple plantar contact points of the robot in the target behavior, wherein the action sequence includes multiple action stages, and the action stages include the theoretical contact states of the multiple plantar contact points. There is at least one plantar release point with a different theoretical contact state between different action stages, and adjacent action stages in the action sequence are divided based on contact state switching events or contact state switching times.

[0038] According to a fifth aspect of an embodiment of the present disclosure, a robot is provided, wherein the robot has a plurality of sole contact points, and the sole of the robot has a plurality of contact sensors, wherein the sole contact point corresponds to at least one contact sensor.

[0039] According to a sixth aspect of an embodiment of the present disclosure, a robot is provided, comprising a memory and a processor, wherein the memory is used to store computer instructions executable on the processor, and the processor is used to implement the robot control method described in the first aspect when executing the computer instructions.

[0040] According to a seventh aspect of an embodiment of the present disclosure, there is provided a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, the method described in the first aspect is implemented.

[0041] The technical solution provided by the embodiments of the present disclosure may have the following beneficial effects:

[0042] The robot control method provided by the disclosed embodiment can control the robot to complete the target behavior through movement according to the action sequence of the target behavior and the actual contact state of the multiple sole contact points by acquiring the actual contact state of the multiple sole contact points. Since the action sequence includes multiple action stages, the action stage includes the theoretical contact state of the multiple sole contact points, and there is at least one sole contact point with different theoretical contact states between different action stages, the action sequence can describe the target movement more accurately and closely, especially greatly improving the anthropomorphism level of the action sequence compared to the case where the sole includes only one contact point; therefore, controlling the robot movement based on the action sequence and the actual contact state of multiple sole contact points can improve the complexity, anthropomorphism level and stability of the robot movement. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0044] Figure 1 is a schematic structural diagram of a robot shown in an exemplary embodiment of the present disclosure;

[0045] Figure 2 is a flow chart of a behavior construction method shown in an exemplary embodiment of the present disclosure;

[0046] Figure 3 is a schematic diagram of the contact points of the sole of the foot shown in an exemplary embodiment of the present disclosure;

[0047] Figure 4 is a schematic diagram of an event-based action sequence shown in an exemplary embodiment of the present disclosure;

[0048] Figure 5 is a schematic diagram of a time-based action sequence shown in an exemplary embodiment of the present disclosure;

[0049] Figure 6 is a schematic diagram of an action sequence of a jumping action shown in an exemplary embodiment of the present disclosure;

[0050] Figure 7 is a schematic diagram of an action sequence of a jumping action in the related art;

[0051] Figure 8 is a flow chart of a robot control method shown in an exemplary embodiment of the present disclosure;

[0052] Fig. 9 is a distribution diagram of contact sensors shown in an exemplary embodiment of the present disclosure;

[0053] Fig.10is a distribution diagram of contact sensors shown in an exemplary embodiment of the present disclosure;

[0054] Fig.11 is a schematic structural diagram of a robot control device shown in an exemplary embodiment of the present disclosure;

[0055] Fig.12 is a structural schematic diagram of a behavior construction device shown in an exemplary embodiment of the present disclosure;

[0056] Fig.13 It is a structural block diagram of a robot shown in an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0057] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0058] The terms used in this disclosure are for the purpose of describing specific embodiments only and are not intended to limit the disclosure. The singular forms of "a", "said" and "the" used in this disclosure and the appended claims are also intended to include plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.

[0059] It should be understood that although the terms first, second, third, etc. may be used in the present disclosure to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present disclosure, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the word "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining".

[0060] In recent years, the technology of robotics has been developing continuously, becoming more and more intelligent and automated, and the richness, stability and flexibility of actions have been improved to varying degrees. Robots can replace users in production and life to perform specific tasks, thus bringing convenience to users. In related technologies, the actions that robots can perform are relatively simple, that is, they are not yet able to perform more complex actions.

[0061] For example, the robot's contact with the outside world is described by the external force applied to a contact point on the sole of the foot, which makes it too simplified and less practical. It cannot be applied to the dynamic behavior of all robots, making it difficult to describe complex dynamic behaviors, and the execution effect is poor or cannot be executed.

[0062] Based on this, at least one embodiment of the present disclosure provides a behavior construction method, which is used to construct an action sequence for a robot, for example, to construct an action sequence for a biped robot (humanoid robot) and a quadruped robot (robot dog). Figure 1 , which takes a bipedal robot as an example to illustrate the structure of the robot. The robot has two legs and two arms, and the legs and arms are provided with drivable joint actuators; the ends of the two legs are provided with two flat feet, which can contact the outside world (such as the ground). By changing the contact state of each foot with the outside world, a variety of dynamic behaviors can be achieved, such as walking, running, jumping, somersaults, stable standing, etc.

[0063] Please refer to the attached Figure 2 , which exemplarily shows the process of the method, including step S201. The method is described below by taking a humanoid robot as an example, and it should be understood that this is not a limitation on the type of robot targeted by the method.

[0064] In step S201, based on the changes in the theoretical contact states of multiple plantar contact points of the robot in the target behavior, an action sequence of the target behavior is constructed, wherein the action sequence includes multiple action stages, and the action stages include the theoretical contact states of the multiple plantar contact points. There is at least one plantar contact point with a different theoretical contact state between different action stages, and adjacent action stages in the action sequence are divided based on contact state switching events or contact state switching times.

[0065] Among them, the method configures multiple sole contact points for the robot, and uses the theoretical contact states of these sole contact points to accurately and closely describe the contact between the sole of the foot and the outside world during the robot's movement. Exemplarily, the multiple sole contact points belong to the foot ends of the robot, and the number of sole contact points at each foot end is the same, that is, these sole contact points are evenly and symmetrically distributed on each foot, which can make the behavior constructed based on this symmetric between different foot ends, and each foot end bears the same amount of action, thereby improving the balance and stability of the robot's movement.

[0066] For a humanoid robot, as the amount of sole contact with the outside world changes, the entire robot system is in different behavioral states (for example, for jumping in place, both feet are in full contact with the ground (or other contact objects) during the take-off phase; and both feet have no contact with the ground (or other contact objects) during the airborne phase). This application introduces the concepts of contact flag (contact_flag) and gait mode (mode). Contact_flag is used to identify whether each sole contact point is in contact with the outside world: 0 indicates no contact, and 1 indicates contact. For example, contact_flag is an array containing elements corresponding to each sole contact point, and each element uses 0 or 1 to indicate whether its corresponding sole contact point is in contact with the outside world; mode includes mode_number and mode_name, which respectively represent the mode number and mode name in the current contact state, and there is a one-to-one correspondence between mode and contact_flag. Next, the attached figure shows the gait mode. Figure 3 Taking the left and right feet shown as having 8 sole contact points as an example, the sole contact point mode and cantact_flag encoding are introduced as follows:

[0067] Please refer to the attached Figure 3 The robot is a bipedal robot, and each foot has four sole contact points distributed at the upper left, lower left, upper right, and lower right. The eight sole contact points of the left and right feet are represented as {LLT, LRT, LLH, LRH, RLT, RRT, RLH, RRH}, respectively, where the first letter represents the left (L) foot or the right (R) foot, the second letter represents the left (L) side or the right (R) side of a single foot, and the third letter represents the toe (Toe) or the heel (Heel) of a single foot.

[0068] contact_flag is defined as an array with 8 elements, the element type is an integer and can only take the value 0 or 1. The 8 elements in the array correspond one-to-one to the above-mentioned 8 sole contact points. When the sole contact point is in contact with the outside world, the corresponding element is 1, and when the sole contact point is not in contact with the outside world, the corresponding element is 0.

[0069] Theoretically, the eight sole contact points can describe 2 8mode. However, considering the walking habits of humans, the following simplification can be made: let the multiple sole contact points be combined into at least one point combination, wherein the point combination includes at least one sole contact point; and the theoretical contact states of multiple sole contact points in the same point combination are always the same. For example, let the two points on the front toes always be in the same contact state, and the two sole contact points on the back heel always be in the same contact state. This is because in the walking habits of humans, the left and right sides of the front toes are generally in contact with or not in contact with the outside world at the same time, and the left and right sides of the back heels are generally in contact with or not in contact with the outside world at the same time.

[0070] After simplification, the left (Left) and right (Right) feet may be in four contact states: non-contact (N), toe contact (T), heel contact (H), and all contact (A). In this way, all the sole contact points of the left and right feet can describe the 2 shown in the following table 4 kind of mode.

[0071] Table 1: Contact status table

[0072]

[0073]

[0074] In the above table, mode_name is named according to the contact status of the left and right feet. For example, LN_RN means that the left and right feet are in the non-contact (N) state, that is, the completely airborne state FLY; LA_RA means that the left and right feet are in the full contact (A) state, that is, both feet are in full contact with the outside world STANCE.

[0075] For humanoid robots, their contact with the outside world often changes with different stages of a behavior, which is reflected in the contact state between each sole contact point and the ground. For example, during walking, each foot will cycle between the support phase and the swing phase, accompanied by the foot landing and leaving the ground. When the contact changes, continuous dynamics alone are not enough to describe the entire behavior process. In other words, some behaviors of humanoid robots inevitably have discrete processes (discrete dynamics). The entire behavior is a hybrid system, which consists of several continuous subsystems and several discrete events, where each continuous subsystem can be called an action phase and each discrete event can be called a switching event.

[0076] Therefore, when describing the target behavior, the occurrence sequence of multiple continuous subsystems in the hybrid system can be mathematically represented to obtain the action sequence of the target behavior. For example, the above gait mode is used to represent the continuous subsystem, and the continuous subsystem can be called an action stage, and each discrete event between the continuous subsystems can be described by the change of the above contact flag (contact_flag). In this way, the attached Figure 4 The event-based action sequence diagram describing the target behavior hybrid system shown in the figure, or the attached Figure 5 The action sequence diagram of the time-based hybrid system describing the target behavior is shown in FIG. Figure 4 The n in the continuous subsystem mode-n is not mode_num, but the sequence number of the sequence. The same mode_num can appear multiple times or even periodically in an action sequence.

[0077] For example, taking the periodic continuous jumping of a humanoid robot as an example, its action sequence diagram is shown in the attached figure. Figure 6 As shown, it includes the cycles of LARA (full foot contact phase), LTRT (forefoot contact phase), LNRN (launch phase), LHRH (rearfoot contact phase), LARA (full foot contact phase), as well as the heel lift switching event between LARA and LTRT, the toe lift switching event between LTRT and LNRN, the heel landing switching event between LNRN and LHRH, and the toe landing switching event between LHRH and LARA.

[0078] In the related art, the contact state between the sole of the foot and the outside world is described by using a contact point, which can only be obtained as shown in the following figure. Figure 7 The action sequence diagram of periodic continuous jumps shown in the figure shows that the action sequence constructed by the present application has the advantages of being more accurate, more in line with the real situation, more anthropomorphic, and more complex than the action sequence constructed in the related art. That is, the behavior construction method provided by the present application can adapt to more complex work; furthermore, robot control based on the action sequence constructed by the present application can improve the complexity and stability of the robot's movement.

[0079] Based on this, at least one embodiment of the present disclosure provides a robot control method, which is used to control legged robots such as biped robots (humanoid robots) and quadruped robots (robot dogs) to complete target behaviors based on the action sequence of the target behaviors.

[0080] The action sequence of the target behavior can be pre-constructed in the following manner: based on the changes in the theoretical contact states of the plurality of sole contact points in the target behavior, the action sequence of the target behavior is constructed, wherein adjacent action stages in the action sequence are divided based on contact state switching events or contact state switching times. The details and methods in the specific construction process may be the same as or different from those provided in the above embodiments.

[0081] The action sequence may include multiple action stages, the action stages include theoretical contact states of the multiple sole contact points, and there is at least one sole contact point with a different theoretical contact state between different action stages. For example, the target behavior includes a jumping behavior, and the action sequence of the target behavior includes a full foot contact stage, a forefoot contact stage, a flight stage, a rear foot contact stage, and a full foot contact stage arranged in sequence.

[0082] Exemplarily, the plurality of sole contact points belong to the foot ends of the robot, and the number of sole contact points of each foot end is the same, that is, these sole contact points are evenly and symmetrically distributed on each foot, which can make the behavior constructed based on this symmetric between different foot ends, and each foot end bears the same amount of action, thereby improving the balance and stability of the robot's movement. Figure 3 For example, the robot is a bipedal robot, and the foot end has four sole contact points distributed at the upper left, lower left, upper right, and lower right.

[0083] Please refer to the attached Figure 8 , which exemplarily shows the process of the method, including step S801 to step S802.

[0084] In step S801, the actual contact states of multiple sole contact points of the robot are obtained, wherein the actual contact states include touching the ground and not touching the ground.

[0085] The sole of the robot has a plurality of contact sensors, wherein (each) of the sole contact points corresponds to at least one of the plurality of contact sensors. Preferably, the number of contact sensors corresponding to each sole contact point is the same. For example, a plurality of contact sensors are distributed between the support plate frame (generally made of metal) and the buffer pad (generally made of rubber) of the foot; the contact sensors may be piezoelectric, switch, etc.; and a plurality of contact sensors may be distributed in a rectangular array on the sole.

[0086] Therefore, this step can obtain the contact parameters collected by the multiple contact sensors, and determine the contact state of (each) the sole contact point according to the contact parameters collected by at least one contact sensor corresponding to (each) the sole contact point. For example, the contact parameters are the linear force and torque collected by the contact sensor, and if the contact force is greater than a preset threshold, the contact state of the corresponding sole contact point is determined to be ground contact, otherwise, the contact state of the corresponding sole contact point is determined to be not ground contact.

[0087] Attach Figure 3 Taking the four sole contact points of the left and right feet as an example, the soles of the feet can be set as shown in the following figure. Fig. 9 There are four contact sensors (each black dot in the figure represents a sensor), and the four contact sensors correspond to four sole contact points one by one. In this case, the contact parameters collected by each contact sensor can be obtained to determine the contact state of each sole contact point.

[0088] Attach Figure 3 Taking the four sole contact points of the left and right feet as an example, the soles of the feet can be set as shown in the following figure. Fig.10 The four groups of contact sensors shown in the figure (each black dot in the figure represents a sensor) each group of contact sensors includes a plurality of contact sensors uniformly distributed in an approximate matrix, and the four groups of contact sensors correspond to four sole contact points one by one. In this case, the contact parameters collected by each contact sensor in each group of contact sensors can be obtained, and the contact parameters of the group of contact sensors can be determined based on the contact parameters collected by all contact sensors in the group (for example, by weighted summation, averaging, etc.), and then the contact state of each sole contact point can be determined based on the contact parameters of each group of contact sensors.

[0089] By setting up multiple contact sensors to determine the contact state corresponding to each sole contact point, the actual state of the sole can be obtained, providing accurate state feedback for the control of the robot, thereby improving the accuracy and complexity of the robot control.

[0090] In addition, if the action sequence of the target behavior is constructed with the following simplifications based on human walking habits: the multiple sole contact points include at least one point combination, wherein the point combination includes at least one sole contact point; in each action stage of the action sequence, the theoretical contact states of multiple sole contact points in the same point combination are the same. Then this step can obtain the actual contact state of any sole contact point in each point combination in the at least one point combination as the actual contact state of the corresponding point combination. This simplifies the process of obtaining the actual contact state, so that its acquisition process is compatible with the simplification of the action sequence.

[0091] In step S802, according to the action sequence of the target behavior and the actual contact states of the plurality of sole contact points, the robot is controlled to complete the target behavior through movement.

[0092] For example, this step can use a variety of optimization control methods based on the action sequence and the actual contact state to achieve trajectory optimization of the target behavior and model predictive control (MPC) to control the robot to complete the target behavior.

[0093] The robot control method provided by the disclosed embodiment can control the robot to complete the target behavior through movement according to the action sequence of the target behavior and the actual contact state of the multiple sole contact points by acquiring the actual contact state of the multiple sole contact points. Since the action sequence includes multiple action stages, the action stage includes the theoretical contact state of the multiple sole contact points, and there is at least one sole contact point with different theoretical contact states between different action stages, the action sequence can describe the target movement more accurately and closely, especially greatly improving the anthropomorphism level of the action sequence compared to the case where the sole includes only one contact point; therefore, controlling the robot movement based on the action sequence and the actual contact state of multiple sole contact points can improve the complexity, anthropomorphism level and stability of the robot movement.

[0094] At least one embodiment of the present disclosure further provides a robot control device. Fig.11 The device comprises:

[0095] An acquisition module 1101 is used to acquire actual contact states of multiple sole contact points of the robot, wherein the actual contact states include ground contact and non-ground contact;

[0096] The control module 1102 is used to control the robot to complete the target behavior through movement according to the action sequence of the target behavior and the actual contact states of the multiple sole contact points, wherein the action sequence includes multiple action stages, and the action stages include theoretical contact states of the multiple sole contact points, and there is at least one sole release point with a different theoretical contact state between different action stages.

[0097] In one embodiment of the present disclosure, the sole of the robot has a plurality of contact sensors, wherein the sole contact point corresponds to at least one contact sensor among the plurality of contact sensors;

[0098] The acquisition module is used for:

[0099] The contact parameters collected by the multiple contact sensors are acquired, and the contact state of the sole contact point is correspondingly determined according to the contact parameters collected by at least one contact sensor corresponding to the sole contact point.

[0100] In one embodiment of the present disclosure, the apparatus further comprises a building block for:

[0101] Based on the changes of the theoretical contact states of the multiple sole contact points in the target behavior, an action sequence of the target behavior is constructed, wherein adjacent action stages in the action sequence are divided based on contact state switching events or contact state switching times.

[0102] In one embodiment of the present disclosure, the plurality of sole contact points include at least one point combination, wherein the point combination includes at least one sole contact point;

[0103] In each action phase of the action sequence, the theoretical contact states of multiple sole contact points in the same point combination are the same;

[0104] The acquisition module is used for:

[0105] The actual contact state of any sole contact point in each point combination in the at least one point combination is obtained as the actual contact state of the corresponding point combination.

[0106] In one embodiment of the present disclosure, the plurality of sole contact points belong to the foot ends of the robot, and the number of sole contact points of each foot end is the same.

[0107] In one embodiment of the present disclosure, the robot is a bipedal robot; and / or the foot end has four sole contact points distributed at the upper left, lower left, upper right, and lower right.

[0108] In one embodiment of the present disclosure, the target behavior includes a jumping behavior, and the action sequence of the target behavior includes a full-foot contact phase, a forefoot contact phase, a flight phase, a rear-foot contact phase, and a full-foot contact phase arranged in sequence.

[0109] At least one embodiment of the present disclosure further provides a behavior construction device. Fig.12 , the device comprises:

[0110] Construction module 1201 is used to construct an action sequence of the target behavior based on the changes in the theoretical contact states of multiple plantar contact points of the robot in the target behavior, wherein the action sequence includes multiple action stages, and the action stages include the theoretical contact states of the multiple plantar contact points. There is at least one plantar contact point with a different theoretical contact state between different action stages, and adjacent action stages in the action sequence are divided based on contact state switching events or contact state switching times.

[0111] At least one embodiment of the present disclosure further provides a robot, the robot having a plurality of sole contact points, the sole of the robot having a plurality of contact sensors, wherein the sole contact points correspond to at least one contact sensor. Figure 1 The humanoid robot shown (the contact sensors on the soles of the feet are not shown).

[0112] At least one embodiment of the present disclosure further provides a robot. Fig.13 , which shows the structure of the robot, the robot includes a memory and a processor, the memory is used to store computer instructions that can be executed on the processor, and the processor is used to control the robot based on the method described in any of the above embodiments when executing the computer instructions.

[0113] At least one embodiment of the present disclosure further provides a computer-readable storage medium having a computer program stored thereon, wherein when the program is executed by a processor, the robot control method described in any of the aforementioned embodiments is implemented.

[0114] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are intended to be exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0115] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A robot control method, characterized in that: The method comprises: Acquire the actual contact status of multiple sole contact points of the robot, wherein the actual contact status includes touching the ground and not touching the ground; According to the action sequence of the target behavior and the actual contact states of the multiple sole contact points, the robot is controlled to complete the target behavior through movement, wherein the action sequence includes multiple action stages, the action stages include theoretical contact states of the multiple sole contact points, and there is at least one sole contact point with a different theoretical contact state between different action stages.

2. The robot control method according to claim 1, characterized in that: The sole of the robot has a plurality of contact sensors, wherein the sole contact point corresponds to at least one contact sensor among the plurality of contact sensors; The obtaining of the actual contact states of the multiple sole contact points of the robot includes: The contact parameters collected by the multiple contact sensors are acquired, and the contact state of the sole contact point is correspondingly determined according to the contact parameters collected by at least one contact sensor corresponding to the sole contact point.

3. The robot control method according to claim 1, characterized in that: The method further comprises: Based on the changes of the theoretical contact states of the multiple sole contact points in the target behavior, an action sequence of the target behavior is constructed, wherein adjacent action stages in the action sequence are divided based on contact state switching events or contact state switching times.

4. The robot control method according to claim 1 or 3, characterized in that: The plurality of sole contact points include at least one point combination, wherein the point combination includes at least one sole contact point; In each action phase of the action sequence, the theoretical contact states of multiple sole contact points in the same point combination are the same; The obtaining of the actual contact states of the multiple sole contact points of the robot includes: The actual contact state of any sole contact point in each point combination in the at least one point combination is obtained as the actual contact state of the corresponding point combination.

5. The robot control method according to claim 1, characterized in that: The plurality of sole contact points belong to the foot ends of the robot, and the number of sole contact points of each foot end is the same.

6. The robot control method according to claim 1 or 5, characterized in that: The robot is a bipedal robot; and / or the foot end has four sole contact points distributed at the upper left, lower left, upper right, and lower right.

7. The robot control method according to claim 6, characterized in that: The target behavior includes a jumping behavior, and the action sequence of the target behavior includes a full-foot contact phase, a front-foot contact phase, a flight phase, a rear-foot contact phase and a full-foot contact phase arranged in sequence.

8. A behavior construction method, characterized in that: The method comprises: Based on the changes in the theoretical contact states of multiple plantar contact points of the robot in the target behavior, an action sequence of the target behavior is constructed, wherein the action sequence includes multiple action stages, and the action stages include the theoretical contact states of the multiple plantar contact points. There is at least one plantar contact point with a different theoretical contact state between different action stages, and adjacent action stages in the action sequence are divided based on contact state switching events or contact state switching times.

9. A robot control device, characterized in that: The device comprises: An acquisition module, used to acquire actual contact states of multiple sole contact points of the robot, wherein the actual contact states include ground contact and non-ground contact; A control module is used to control the robot to complete the target behavior through movement according to the action sequence of the target behavior and the actual contact states of the multiple sole contact points, wherein the action sequence includes multiple action stages, the action stages include theoretical contact states of the multiple sole contact points, and there is at least one sole release point with a different theoretical contact state between different action stages.

10. A behavior construction device, characterized in that: The device comprises: A construction module is used to construct an action sequence of the target behavior based on changes in the theoretical contact states of multiple plantar contact points of the robot in the target behavior, wherein the action sequence includes multiple action stages, and the action stages include the theoretical contact states of the multiple plantar contact points. There is at least one plantar release point with a different theoretical contact state between different action stages, and adjacent action stages in the action sequence are divided based on contact state switching events or contact state switching times.

11. A robot, characterized in that: The robot has a plurality of sole contact points, and the sole of the robot has a plurality of contact sensors, wherein each sole contact point corresponds to at least one contact sensor.

12. A robot, characterized in that: The robot comprises a memory and a processor, wherein the memory is used to store computer instructions executable on the processor, and the processor is used to implement the method according to any one of claims 1 to 8 when executing the computer instructions.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images