Deformable mobile vehicle

Abstract

The deformable moving trolley comprises a variable geometry truss, a moving execution assembly and a cooperative control assembly, the moving execution assembly is connected with the variable geometry truss, the variable geometry truss comprises a plurality of independent rod pieces with the lengths capable of changing actively, and the rod pieces are connected based on connecting nodes; the connecting node can realize movable connection of at least two rotational degrees of freedom or fixed connection of rigid locking, at least part of the connecting node is in movable connection, and the rod piece with the length capable of actively changing can actively extend or retract according to a control instruction. The cooperative control assembly is used for controlling the change of the geometric configuration of the variable geometry truss and the movement of the movement execution assembly; and the moving execution assembly bears the variable geometry truss to move according to a specified mode under the control of the cooperative control assembly. According to the invention, the flexible change of the form can be realized through the extension and retraction of the rod piece and the rotation of the rod piece around the node, so that the mobile vehicle can efficiently adapt to various complex environments, and the limitation of a traditional robot system is overcome.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent mechanical control, and particularly relates to a deformable mobile vehicle. BACKGROUND

[0002] With the rapid development of robot technology, its application scenarios have gradually expanded to complex terrain environments, such as stairs, slopes, rugged roads and the like. However, the existing mobile robots, especially wheeled and tracked robots, are designed to be suitable only for flat ground, and have very limited obstacle-crossing ability when facing stairs, slopes or rugged roads. Especially for stairs and other obstacles with large height differences, wheeled robots often cannot pass normally.

[0003] In addition, the suspension system and tire design of traditional wheeled robots cannot effectively adjust the vehicle body posture, resulting in poor stability, especially in steep slopes or irregular terrain environments, and are prone to rollover or loss of control.

[0004] In summary, the existing wheeled robots cannot provide sufficient stability and flexibility in complex terrain.

[0005] At present, there is also a quadruped robot which can adapt to irregular ground and has certain obstacle-crossing ability, but its implementation structure is complex, the cost is high, the manufacturing difficulty is great, and the energy consumption is high. At the same time, the corresponding quadruped robot is still difficult to maintain stability in some complex environments, especially in relatively extreme terrain, and still faces not small challenges.

[0006] In view of this, the present application is proposed. SUMMARY

[0007] The present application aims to provide a deformable mobile vehicle to solve the technical problem that the existing robots cannot adapt to complex environments.

[0008] The present application aims to achieve the following technical solutions:

[0009] A deformable mobile vehicle, comprising a variable geometry truss, a mobile execution assembly and a cooperative control assembly, wherein:

[0010] The variable geometry truss comprises a plurality of independent rods connected by connecting nodes, and some or all of the rods are rods with active length change; the connecting node comprises at least two connecting ends, each of which is connected with one end of the rod; the connecting mode between the connecting end and the rod is configured as movable connection or fixed connection, wherein at least part of the connecting ends are connected with one end of the rod through movable connection with at least two rotational degrees of freedom; the rod with active length change is a rod which can actively perform elongation or shortening operation according to control instructions;

[0011] The mobile execution assembly is connected with the variable geometry truss and the cooperative control assembly, and is used to carry the variable geometry truss to move in a specified manner under the control of the control instructions of the cooperative control assembly;

[0012] The cooperative control assembly is electrically connected with the rod with active length change and the mobile execution assembly, and is used to generate control instructions and send them to the rod with active length change and the mobile execution assembly, so as to control the change of the geometric configuration of the variable geometry truss and the movement of the mobile execution assembly.

[0013] When the connecting mode is movable connection, the connecting node comprises at least one of a spherical joint, a universal joint, a three-rotation-axis co-point joint, a flexible spherical hinge or a multi-axis composite hinge; when the connecting mode is fixed connection, the connecting node comprises a rigid locking structure; the rod with active length change comprises at least one of an electric push rod, a linear actuator, a hydraulic cylinder, an air pressure rod or a lead screw nut mechanism driven by a servo motor; and the mobile execution assembly and the variable geometry truss adopt detachable or reconfigurable connecting structure, and each independent rod and connecting node of the variable geometry truss adopt detachable or reconfigurable connecting structure.

[0014] The deformable mobile vehicle further comprises a perception assembly for obtaining the internal state of the deformable mobile vehicle and / or the environmental data around the mobile vehicle, and providing them as inputs to the cooperative control assembly for generating corresponding control instructions; wherein,

[0015] The internal state information comprises one or more of the length of the rod with active length change, the length change rate, the driving parameter, the force at both ends or the speed, acceleration and attitude of each connecting node; the driving parameter comprises one or more of driving current, voltage, air pressure or hydraulic pressure value;

[0016] The environmental data around the vehicle comprises one or more of the environmental geometry, the spatial distribution of obstacles and the scene semantic recognition information obtained by the visual sensor.

[0017] The surrounding environment data is obtained by an environment perception component, and the environment perception component comprises a data acquisition module and a data processing module, wherein:

[0018] The data acquisition module is configured to acquire the environment data by at least one of a three-primary-color RGB camera, a depth camera, a laser radar or a millimeter wave radar as an environment perception sensor;

[0019] The data processing module is configured to process the raw data acquired by the data acquisition module to generate a terrain representation of the surrounding environment of the deformable mobile vehicle.

[0020] The data processing module adopts a simultaneous localization and mapping (SLAM) technology to construct a terrain map representing the surrounding environment; or the data processing module directly performs real-time analysis on the sensor data at the current time, generates a corresponding local terrain representation by calculating local features, and obtains the terrain features in front in real time.

[0021] The cooperative control component comprises an intelligent decision module configured to obtain a cooperative adjustment sequence capable of optimizing a preset performance index according to the terrain representation generated by the perception component and the internal state information of the vehicle, and ensure that the cooperative adjustment sequence always satisfies a preset physical constraint; and a unified representation of the core component of the intelligent decision module comprises:

[0022] The state space as input is defined as a comprehensive state representation, which at least comprises environment data, vehicle body posture data and complete internal state information of the mobile vehicle; and the complete internal state information comprises a high-dimensional configuration state of the variable geometry truss and a state of the mobile execution component;

[0023] The action space as output comprises cooperative control instructions for a plurality of length-adjustable rod members and control instructions for the mobile execution component;

[0024] The performance index optimization target is configured to quantify the comprehensive operation performance of the mobile vehicle;

[0025] The physical constraint is configured to reflect the physical and environmental limitations of the mobile vehicle.

[0026] The deformable mobile vehicle further comprises an end effector component installed on the variable geometry truss and / or the mobile execution component, configured to physically interact with the external environment and / or target object under the control of the cooperative control component to perform operation tasks.

[0027] The end effector component comprises:

[0028] The grabbing and operation execution unit comprises a grabbing component and / or an operation component, and is used for executing grabbing and / or a predetermined operation action on the target;

[0029] And / or,

[0030] The perception and interaction unit is used for closed-loop feedback control based on the execution of the grabbing and operation execution unit, so as to control adjustment of the execution of the grabbing and / or the predetermined operation action.

[0031] The deformable mobile vehicle further comprises a structural state feedback adjustment assembly, which comprises:

[0032] The state monitoring unit is used for monitoring the telescopic amount and / or stress condition of the rod in real time based on the self-internal state information acquired by the perception assembly;

[0033] The adjustment controller is used for outputting a compensation control signal to adjust the structural form of the variable geometry truss to realize stress redistribution or form error correction when the monitored data deviates from the preset safety range or target parameter.

[0034] The adjustment controller comprises:

[0035] The storage module is used for pre-storing the safety stroke range and / or safety stress range of each rod, and pre-storing the allowable form error threshold of the actual configuration of the variable geometry truss relative to the target configuration;

[0036] The control module is used for reading the telescopic amount and / or stress condition collected by the state monitoring unit, and determining that the data deviates from the preset safety range or target parameter when the current telescopic length or stress of any rod exceeds the corresponding safety range, and / or the deviation between the actual configuration of the variable geometry truss determined according to the current length of all rods and the target configuration exceeds the allowable form error threshold, and generating a compensation control signal according to the type and amplitude of the deviation to adjust the target telescopic length of one or more rods with actively changeable length.

[0037] Compared with the prior art, the deformable mobile vehicle can effectively improve the adaptability of the mobile vehicle as a robot in a complex environment. Specifically, the flexible change of the mobile vehicle body form can be realized through the telescopic amount of the rod and the rotation of the rod around the node with at least two rotation degrees of freedom, so that the mobile vehicle can efficiently adapt to various complex environments and overcome the limitations of traditional robot systems. For example, the corresponding mobile vehicle can stably move and run on complex terrains such as stairs and slopes, and can be self-adaptively adjusted to complete different complex motion modes. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed in the embodiment description. Obviously, the drawings in the following description only show some of the embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without any creative effort.

[0039] Figure 1 A structural schematic diagram of the deformable mobile vehicle provided by the embodiment of the present application;

[0040] Figure 2 A principle structural schematic diagram of the deformable mobile vehicle provided by the embodiment of the present application. DETAILED DESCRIPTION

[0041] The technical solutions in the embodiments of the present application will be described clearly and completely in the following with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments only show some of the embodiments of the present application, but not all the embodiments of the present application, which does not constitute a limitation to the present application. Based on the embodiments of the present application, all the other embodiments obtained by those skilled in the art without any creative effort belong to the protection scope of the present application.

[0042] Firstly, the terms possibly used in the present application are explained as follows:

[0043] The term "and / or" means either of the two or both can be realized, for example, X and / or Y means three cases including "X" or "Y" or "X and Y".

[0044] The terms "include", "contain", "have", "possess" or other similar semantic descriptions should be interpreted as non-exclusive inclusion. For example, including a technical feature element (such as raw materials, components, ingredients, carriers, dosage forms, materials, sizes, parts, components, mechanisms, devices, steps, processes, methods, reaction conditions, processing conditions, parameters, algorithms, signals, data, products or articles, etc.) should be interpreted as not only including the explicitly listed technical feature element, but also including other technical feature elements not explicitly listed in the art.

[0045] The term "consisting of" means excluding any technical feature element not explicitly listed. If this term is used in the claims, the term will make the claim closed, so that it does not contain technical feature elements other than the explicitly listed technical feature elements, but the conventional impurities related thereto are excluded. If the term only appears in a certain clause of the claim, it only limits the elements explicitly listed in the clause, and the elements recorded in other clauses are not excluded from the overall claim.

[0046] The term "parts by mass" is used to express the mass ratio relationship between components, for example, if it is described that X component is x parts by mass and Y component is y parts by mass, it means that the mass ratio of X component to Y component is x:y; 1 part by mass can represent any mass, for example, 1 part by mass can represent 1 kg or 3.1415926 kg, etc. The sum of the parts by mass of all components does not necessarily equal 100 parts, and can be greater than, less than, or equal to 100 parts. Unless otherwise specified, the parts, ratios, and percentages described herein are by mass.

[0047] Unless specifically defined or limited otherwise, the terms "mounting", "connected", "connection", "fixed", and the like, are used broadly and encompass both direct and indirect mounting, connection, and fixation, as well as fixed or detachable mounting, connection, and fixation, mechanical connection, electrical connection, direct connection, indirect connection, and internal communication between two elements. Those skilled in the art can understand the specific meaning of the above terms in this document according to the specific circumstances.

[0048] When a concentration, temperature, pressure, size, or other parameter is expressed in a numerical range, the numerical range should be understood to specifically disclose all ranges formed by any pair of upper limit values, lower limit values, and preferred values within the numerical range, regardless of whether the range is explicitly recited; for example, if the numerical range "2-8" is recited, the numerical range should be interpreted to include ranges such as "2-7", "2-6", "5-7", "3-4 and 6-7", "3-5 and 7", "2 and 5-7", etc. Unless otherwise specified, the numerical ranges recited herein include all integers and fractions within the numerical range, as well as the end values.

[0049] The terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", and the like, indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are used only for the convenience of description and simplification of description, and do not mean that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be construed as limiting the present document.

[0050] A deformable mobile vehicle capable of highly adapting to complex terrain provided by the present application will be described in detail below. The contents not described in detail in the embodiments of the present application belong to the prior art known to those skilled in the art. Unless otherwise specified, the embodiments of the present application are carried out under conventional conditions in the art or under conditions recommended by the manufacturer. Unless otherwise specified, the reagents or instruments used in the embodiments of the present application are conventional products that can be obtained by commercial purchase.

[0051] The application discloses a deformable mobile vehicle with high adaptability to complex terrain, and belongs to the technical fields of robots, mechatronics and vehicle engineering, and the like. The deformable mobile vehicle can be moved through the combination of a corresponding variable geometry truss structure and a mobile execution component (such as a wheeled driving system). The variable geometry truss structure can change its shape through the combination of the extension and retraction of the rods included in the variable geometry truss and the rotation of the rods around the nodes with at least two rotational degrees of freedom. The deformable mobile vehicle can also perform additional functions through the addition of a corresponding end effector component (such as a mechanical arm). Furthermore, based on the corresponding shape changes and the movement of the whole vehicle, the deformable mobile vehicle can be applied to multiple fields, especially in application scenarios that require the crossing of complex terrain, such as stairs, rugged roads and other complex terrain, and application fields such as intelligent robots, disaster rescue, medical transportation, building inspection, military applications and the like. The deformable mobile vehicle can not only effectively solve the stability and adaptability problems of existing mobile robots in complex terrain, but also has strong market prospects and wide application potential.

[0052] The deformable mobile vehicle provided by the embodiment of the application can be used in multiple fields, especially in application scenarios that require the crossing of complex terrain, such as stairs, rugged roads and other complex terrain, and application fields such as intelligent robots, disaster rescue, medical transportation, building inspection, military applications and the like. Figure 1 Figure 2 The deformable mobile vehicle provided by the embodiment of the application can be used in multiple fields, especially in application scenarios that require the crossing of complex terrain, such as stairs, rugged roads and other complex terrain, and application fields such as intelligent robots, disaster rescue, medical transportation, building inspection, military applications and the like.

[0053] ​The variable geometry truss comprises a plurality of members 2 and a plurality of connecting nodes 1 for connecting the members, that is, a plurality of independent members 2 connected by connecting nodes 1, at least part (that is, all or part) of the plurality of independent members being a length-actively-changeable member, and the members being connected based on the connecting nodes 1 to enable the geometry (that is, the form) of the vehicle body formed by the variable geometry truss to change; the connecting node 1 comprises at least two connecting ends, each of which is connected to one end of the member as a connecting member; the connecting mode between the connecting end and the member is configured as movable connection or fixed connection, wherein at least part (that is, all or part) of the connecting ends are connected to one end of the member through movable connection with at least two rotational degrees of freedom; specifically, when the connecting mode is movable connection, the connecting node can be at least one of a spherical joint, a universal joint, a three-rotation-axis co-point joint, a flexible spherical hinge, or a multi-axis composite hinge, etc. multi-degree-of-freedom connecting joint; it should be noted that the "connection mode with at least two rotational degrees of freedom" in the present application can be selected according to the needs of cable arrangement or torsional stiffness in specific implementation. For example, in one embodiment, a spherical hinge is used to allow the member to perform three-dimensional motion including rotation around its own axis relative to the node; in another embodiment, a cross shaft universal joint or a limited hinge structure is used to allow the member to swing in multiple directions in space (that is, with two rotational degrees of freedom), but limit its rotation around its own axis; when the connecting mode is fixed connection, the connecting node can be a rigid locking structure;

[0054] The length-actively-changeable member is a member that can actively perform elongation or shortening operation according to an external control signal; for example, the length-actively-changeable member can be but is not limited to an actuator integrated with a driving source, which can itself actively perform elongation or shortening action according to an external control signal; specifically, as a non-limiting example, the length-actively-changeable member can be embodied as at least one of an electric push rod, a linear actuator, a hydraulic cylinder, a pneumatic rod, or a lead screw nut mechanism driven by a servo motor, etc.

[0055] The mobile execution assembly 3 is connected with the variable geometry truss and also connected with the cooperative control assembly, for carrying the variable geometry truss to move in a specified manner under the control of the control instruction of the cooperative control assembly; correspondingly, the mobile execution assembly can be but is not limited to a plurality of wheels arranged on the variable geometry truss;

[0056] The cooperative control component is electrically connected with the length-actively-changeable rod member and the moving execution component, and is configured to generate control instructions to cooperatively control the length-actively-changeable rod member to adjust the length and change the geometric configuration of the variable-geometry truss, and to control the moving execution component to move in a specified manner; specifically, the cooperative control component serves as the control core of the entire deformable mobile vehicle, and can be configured to, on one hand, cooperatively control the length adjustment of the plurality of length-actively-changeable rod members by sending the generated control instructions to different rod members to cooperatively control the overall geometric configuration of the variable-geometry truss; on the other hand, the moving execution component can be controlled to move according to actual needs by corresponding control instructions, so as to control the mobile vehicle to enter a specified position or region to perform a corresponding task.

[0057] In the deformable mobile vehicle, detachable or reconfigurable connection structures can be used between the moving execution component and the variable-geometry truss, and between each independent rod member and the connecting node of the variable-geometry truss, so that the vehicle body of the mobile vehicle and / or the wheels serving as the moving execution component can be quickly detached or reconfigured, thereby forming a configuration form of the mobile vehicle with different numbers of nodes or configurations to meet the customization needs of different application scenarios, such as medical transportation, building inspection, environmental monitoring, and crossing stairs, obstacles, etc. in scenarios with large changes in height, such as stairs, and flexible movement on terrains with large changes in height, which has excellent performance; that is, the structural configuration of the corresponding mobile vehicle can be arbitrarily assembled and deformed according to actual needs, so that the corresponding mobile vehicle can not only be dynamically configured after assembly according to needs, but also can be assembled into any configuration during assembly according to needs, so that the structural configuration of the entire deformable mobile vehicle is more complex and diverse, and the mobile vehicle can meet the needs of more application scenarios, for example, assembling a mobile vehicle with any size according to the application scenario, and then the assembled mobile vehicle can be moderately deformed according to needs to further adapt to the scene working needs.

[0058] Optionally, as shown in Figure 2 The deformable mobile vehicle provided by the embodiment of the present application can further include a perception component and an end effector component, etc.

[0059] (1) Perception component

[0060] The perception component can be used to obtain the internal state of the deformable mobile vehicle and / or the environmental data around the mobile vehicle, and provide it as input to the cooperative control component for generating corresponding control instructions to control the deformation or movement process of the mobile vehicle, wherein:

[0061] The internal state information of the deformable mobile vehicle itself can include, but is not limited to, one or more of the following: the length of each length-variable rod, the length change rate, the driving parameter, the force at both ends, the direction, or one or more of the speed, acceleration, and attitude of each connection node, etc. The driving parameter can include, but is not limited to, one or more of the driving current, voltage, gas pressure, or liquid pressure value.

[0062] The surrounding environment data includes one or more of the following: the environment geometry, the spatial distribution of obstacles, and the scene semantic recognition information obtained by a laser radar, a depth camera, or other visual sensors, etc.

[0063] Furthermore, the above-mentioned cooperative control component can also receive and process the internal state information of the deformable mobile vehicle itself to generate corresponding intelligent decisions and then generate corresponding control instructions, and issue the control instructions to the length-variable rod and / or the mobile execution component, so as to realize adaptive shape adjustment and path planning of the mobile vehicle based on environmental feedback.

[0064] Further, the perception component can include a data acquisition module and a data processing module, wherein:

[0065] The data acquisition module is configured to acquire the surrounding environment data through at least one environmental perception sensor such as an RGB (three primary colors) camera, a depth camera, a laser radar, or a millimeter wave radar.

[0066] The data processing module (also referred to as a terrain representation and understanding module) is configured to process the raw data acquired by the data acquisition module to generate a terrain representation of the surrounding environment of the deformable mobile vehicle. The terrain representation can include, but is not limited to, the following: the passable area, the position and geometric properties (such as height, slope, roughness) of the obstacle, or the semantic information of the environment.

[0067] In one embodiment, the data processing module can use a simultaneous localization and mapping (SLAM) technique to construct a long-term, global terrain map representing the surrounding environment, such as a terrain elevation map. Alternatively, in another embodiment, the data processing module can not rely on a global map, but directly analyze the sensor data (such as laser radar point cloud) at the current time, calculate local slope, flatness, and height mutation, etc. to generate a local and timely terrain representation, so as to obtain the terrain features in front of the vehicle in real time and realize rapid response to the terrain in front of the vehicle. In the implementation process, those skilled in the art can select any existing SLAM or point cloud processing method to implement the data processing module according to the specific application requirements, and the present application does not limit this.

[0068] It needs to be further explained that in the deformable mobile vehicle provided by the embodiment of the application, the cooperative control component can include an intelligent decision module in the specific implementation process, which is used to obtain a cooperative adjustment sequence capable of optimizing a preset performance index according to the terrain representation generated by the perception component and the internal state information of the mobile vehicle, and ensure that the cooperative adjustment sequence always satisfies a preset physical constraint; that is, the cooperative control component receives and processes the state information (including the internal state information of the mobile vehicle and the surrounding environment data) provided by the perception component to generate a control instruction sequence for cooperatively controlling the length-actively-changeable rod members and the mobile execution component, so as to realize adaptive form adjustment and path planning of the mobile vehicle based on environmental feedback.

[0069] That is, the intelligent decision module included in the cooperative control component can be configured to solve a unified cooperative control optimization problem; the optimization problem aims to calculate or deduce a cooperative adjustment sequence (as an action output) capable of optimizing a preset performance index according to the terrain representation (as an external environment input) generated by the perception component and the internal state information (as an internal state input) of the mobile vehicle, while ensuring that the sequence always satisfies a series of preset physical constraints.

[0070] Specifically, the core component of the cooperative control optimization problem can be uniformly represented as:

[0071] State space (input): that is, the state space as input, which is defined as a comprehensive state representation, and the comprehensive state representation at least includes: the external environment data (such as the terrain representation); the vehicle body posture data (such as the IMU data); and the complete internal state information of the mobile vehicle. The complete internal state information preferably includes: the high-dimensional configuration state of the variable geometry truss (for example, the current telescopic length, telescopic speed and / or force at both ends of the length-actively-changeable rod member), and the state of the mobile execution component (for example, the current rotating speed, rotating angle, ground contact sensing signal or driving motor / torque information of each wheel);

[0072] Action space (output): that is, the action space as output, which includes the cooperative adjustment instruction for the plurality of length-actively-changeable rod members (for example, a high-dimensional action vector including the target telescopic length or target output force of the plurality of rod members), and the motion instruction for the mobile execution component (for example, the target rotating speed or target torque of each wheel);

[0073] Optimization target (performance index): that is, the performance index optimization target, which is configured to quantify the comprehensive performance of the mobile vehicle, for example, maximizing the stability (such as the ZMP or SSM margin), minimizing the task tracking error (such as the target speed, navigation path and truss configuration), and minimizing the energy consumption;

[0074] Physical constraints: configured to reflect the physical and environmental limitations of the deformable mobile vehicle, e.g., kinematic and dynamic constraints of the linkages (travel, velocity, force limits), terrain contact constraints, and collision avoidance constraints.

[0075] In particular, the intelligent decision module can solve the above-mentioned same cooperative control optimization problem in one of at least the following three ways:

[0076] Way one: model-based control, in one embodiment (e.g., using model predictive control MPC), the module is configured to explicitly define the “optimization objective” as a pre-set cost function (e.g., a weighted sum of the aforementioned stability, task tracking, energy consumption indicators), and to take the “physical constraints” as boundary conditions of a nonlinear programming. The module solves the optimization problem of such constraints online in real time in each control period to obtain the cooperative adjustment sequence;

[0077] Way two: learning-based control, in another embodiment (e.g., using deep reinforcement learning DRL), the module is configured to implicitly design the “optimization objective” as a reward function (e.g., reward stability, reward tracking, punish energy consumption). The module learns a neural network (decision policy) through offline training in a physical environment or a simulation environment (e.g., using the PPO algorithm). Once the network is trained, it can online and efficiently map the input of the high-dimensional “state space” to the output of the “action space” (i.e., the cooperative adjustment sequence); the mapping relationship has been optimized in the training by maximizing the reward function, thereby implicitly learning how to meet the “physical constraints” (e.g., by punishing collisions or overruns);

[0078] Way three: hybrid-based control, in yet another embodiment, the module can combine the strategies of the “model-based control” and the “learning-based control” to achieve complementary advantages.

[0079] In one example, the “learning-based control” (e.g., DRL) can be used to generate a high-level policy or reference trajectory (e.g., to decide when to lift and when to move forward), while the “model-based control” (e.g., MPC) is used to solve the specific, bottom-level cooperative adjustment sequence that meets the physical constraints to implement the high-level policy online;

[0080] In another example, the “model-based control” (e.g., MPC) serves as the main controller, but the dynamic model or cost function parameters it relies on can be obtained offline through the “learning-based control” method or online self-adaptation;

[0081] In yet another example, the learning-based control (e.g., a DRL policy network) generates a raw adjustment sequence as the primary controller, which is sent to a model-based safety verification module (e.g., a constraint checker based on MPC) for verification or revision to ensure that it strictly satisfies the physical constraints before being executed.

[0082] (2) End effector assembly

[0083] The end effector assembly is mounted on the variable geometry truss and / or mobile execution assembly, and is controlled by the cooperative control assembly to physically interact with the external environment and / or target object to perform operational tasks; for example, it can be but not limited to a mechanical arm mounted on the variable geometry truss, etc.

[0084] The end effector assembly is also controlled by the cooperative control assembly to physically interact with the external environment or target object to perform specific operational tasks; the end effector assembly is directly or indirectly mounted on specific nodes, members and / or mobile execution assemblies of the variable geometry truss, and its specific functions and forms can be configured according to task requirements;

[0085] Further, as a non-limiting example, the end effector assembly can include one or more of the following units:

[0086] Gripping and manipulation unit: including gripping components and / or manipulation components for performing gripping and / or predetermined operation actions on targets; it can be a mechanical gripper, dexterous hand or gripper, etc. for gripping and carrying objects, etc.; the corresponding operation components can be integrated screwdrivers, welding guns, drill bits or detection probes, etc. for performing assembly, maintenance or detection tasks, etc.

[0087] Sensing and interaction unit: for closed-loop feedback control based on the execution of the gripping and operation execution unit to control the adjustment of the execution of the gripping and / or predetermined operation actions; for example, equipped with a camera, force / torque sensor or tactile sensor at the end, so as to be used for local, high-precision closed-loop feedback control of operation tasks, thereby improving the accuracy and safety of operation.

[0088] The deformable mobile vehicle provided in the embodiment of the present application can further comprise a wireless remote control module, for example, a Wi-Fi communication mode can be adopted, and the coordinated control component is remotely controlled in motion path, running speed and variable geometry truss structure adjustment state based on a user datagram protocol (UDP); specifically, the wireless remote control module can realize remote control based on the UDP protocol, for example, an operator can establish a wireless local area network connection with a vehicle-mounted Wi-Fi module through a mobile control terminal, and real-time remote control of the vehicle motion path, running speed and variable geometry truss structure adjustment state can be realized within a certain distance range; in the deformable mobile vehicle, corresponding control instructions can be sent to the vehicle-mounted embedded control unit in the form of structured datagrams through UDP, the mobile vehicle analyzes the received data and responds quickly, and flexible control and adaptive structure adjustment under variable terrain conditions are ensured.

[0089] In the embodiment of the present application, the corresponding deformable mobile vehicle can further comprise a structure state feedback adjustment component, which specifically can comprise a state monitoring unit and an adjustment controller, wherein:

[0090] The state monitoring unit is used for monitoring the telescopic amount and / or stress of the rod based on the self-internal state information obtained by the perception component in real time;

[0091] The adjustment controller is used for outputting a compensation control signal to adjust the structure form of the variable geometry truss when the monitored data deviates from the preset safety range or target parameter;

[0092] Specifically, the adjustment controller can comprise:

[0093] The storage module is used for pre-storing the safe stroke range and / or safe stress range of each rod, for example, the telescopic length of each rod should be limited between the corresponding minimum length and maximum length, and the axial stress should be limited within the material allowable stress range; at the same time, the allowable form error threshold of the actual configuration of the variable geometry truss relative to the target configuration can also be pre-set and stored;

[0094] The control module is used for adjusting the controller to read the parameters such as telescopic length and stress collected by the state monitoring unit periodically during operation, when it is detected that the current telescopic length or stress of any rod exceeds the corresponding safety range, and / or the deviation between the actual truss configuration calculated according to the current lengths of all rods and the target configuration exceeds the preset configuration error threshold, it is determined that the preset safety range or target parameter is deviated, and after the deviation is determined, the adjusting controller can generate a compensation control signal according to the type and amplitude of the deviation to adjust (usually fine-tune) the target telescopic length of one or more rods with active length change: for example, when it is detected that the stress of a rod exceeds the upper limit, the rod can be issued with a correction instruction to shorten or lengthen, and the target length of the adjacent rod connected thereto is adjusted synchronously according to the current support condition, so as to realize stress redistribution while maintaining the vehicle body support polygon and overall stability; when it is detected that the overall configuration deviates from the target configuration, the target length of the plurality of rods can be adjusted in a small amplitude in the opposite direction according to the direction of the configuration error, so as to reduce the configuration error.

[0095] The above compensation process can be realized in a closed-loop control mode, for example, the target length of each rod is iteratively corrected according to a proportional-integral-derivative (PID) control law or other linear / nonlinear control law in each control period, until the monitored parameters are within the preset safety range.

[0096] As can be seen from the above, the deformable mobile vehicle provided by the present application can realize flexible change of the configuration through the telescopic length of the rods and the rotation of the rods around the nodes, so that the mobile vehicle can efficiently adapt to various complex environments and overcome the limitations of traditional robot systems. For example, the corresponding mobile vehicle can stably move and run on complex terrains such as stairs and slopes, and can adaptively adjust to complete different complex motion modes. Moreover, the deformable mobile vehicle based on the variable geometry truss structure can adjust the vehicle body posture under complex terrains, realize adjustment and change of different configurations, and provide stable support.

[0097] The above description is only the preferred embodiment of the present application, but the protection scope of the present application is not limited thereto, any changes or replacements within the technical range disclosed by the present application can be easily thought by those skilled in the art, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims. The information disclosed in the background section of the present application is only intended to deepen the understanding of the overall background of the present application, and should not be regarded as acknowledging or implying in any form that the information constitutes prior art known by those skilled in the art.

Claims

1. A transformable mobility vehicle, characterized in that, The application relates to a variable geometry truss, a moving execution assembly and a cooperative control assembly, wherein: The variable geometry truss comprises a plurality of independent rods connected through connecting nodes, and part or all of the rods are rods with a length capable of being actively changed; the connecting nodes comprise at least two connecting ends, each of which is connected with one end of the rod; the connecting mode between the connecting end and the rod is configured as movable connection or fixed connection, wherein at least part of the connecting ends are connected with one end of the rod through movable connection with at least two rotation degrees of freedom; the rod with a length capable of being actively changed is a rod capable of actively performing elongation or shortening operation according to control instructions; The moving execution assembly is connected with the variable geometry truss and the cooperative control assembly, and is used for carrying the variable geometry truss to move in a specified mode under the control of control instructions of the cooperative control assembly; The cooperative control assembly is electrically connected with the rod with a length capable of being actively changed and the moving execution assembly, and is used for generating control instructions and sending the control instructions to the rod with a length capable of being actively changed and the moving execution assembly so as to control the change of the geometric configuration of the variable geometry truss and the movement of the moving execution assembly. When the connecting mode is movable connection, the connecting node comprises at least one of a spherical joint, a universal joint, a three-rotation-axis co-point joint, a flexible spherical hinge or a multi-axis composite hinge; when the connecting mode is fixed connection, the connecting node comprises a rigid locking structure; the rod with a length capable of being actively changed comprises at least one of an electric push rod, a linear actuator, a hydraulic cylinder, an air pressure rod or a screw nut mechanism driven by a servo motor; and detachable or recombination type connecting structures are adopted between the moving execution assembly and the variable geometry truss, and between each independent rod and the connecting node of the variable geometry truss.

2. The transformable mobility vehicle of claim 1, wherein, A perception assembly is further included, which is used for acquiring self-internal state and / or surrounding environment data of the deformable moving vehicle and providing the data as input to the cooperative control assembly for generating corresponding control instructions; wherein, 3. The transformable mobility vehicle of claim 1 or 2, wherein, The self-internal state information comprises one or more of the following: the extension length, the extension change rate, the driving parameter, the force size and direction of the two ends of each rod with a length capable of being actively changed or one or more of the speed, acceleration and attitude of each connecting node; the driving parameter comprises one or more of the driving current, voltage, air pressure or hydraulic value; The surrounding environment data comprises one or more of the following: the environment geometric structure, the spatial distribution of obstacles and the scene semantic recognition information acquired through a visual sensor. The surrounding environment data is obtained through an environment perception assembly, and the environment perception assembly comprises a data acquisition module and a data processing module, wherein:

4. The transformable mobility vehicle of claim 3, wherein, The data acquisition module is used for acquiring the environment data through at least one of a three-primary-color RGB camera, a depth camera, a laser radar or a millimeter wave radar as an environment perception sensor; ​ The data processing module is configured to process the raw data acquired by the data acquisition module to generate a terrain representation of the surroundings of the deformable mobile vehicle.

5. The transformable mobility vehicle of claim 4, wherein, The data processing module adopts a simultaneous localization and mapping (SLAM) technique to construct a terrain map representing the surroundings, or directly analyzes the sensor data at the current time in real time to generate a corresponding local terrain representation by calculating local features to obtain the terrain features in front of the vehicle in real time.

6. The transformable mobility vehicle of claim 3, wherein, The cooperative control component includes an intelligent decision module configured to obtain a cooperative adjustment sequence that optimizes a preset performance index based on the terrain representation generated by the perception component and the internal state information of the vehicle, and ensure that the cooperative adjustment sequence always satisfies a preset physical constraint. The unified representation of the core component of the intelligent decision module includes: an input state space defined as a comprehensive state representation including at least environmental data, vehicle body posture data, and complete internal state information of the vehicle, and the complete internal state information including high-dimensional configuration state of the variable geometry truss and state of the mobile execution component; an output action space defined as cooperative control instructions for the plurality of length-adjustable rods and control instructions for the mobile execution component; a performance index optimization target configured to quantify the comprehensive operation performance of the vehicle; a physical constraint configured to reflect the physical and environmental limitations of the vehicle.

7. The transformable mobility vehicle of claim 1 or 2, wherein, The end effector component is installed on the variable geometry truss and / or the mobile execution component, and is configured to physically interact with the external environment and / or target object under the control of the cooperative control component to perform operation tasks.

8. The transformable mobility vehicle of claim 7, wherein, The end effector component includes: a grasping and operation execution unit including a grasping component and / or an operation component for performing grasping and / or predetermined operation actions on a target; and / or, a perception and interaction unit configured to perform closed-loop feedback control based on the execution of the grasping and operation execution unit to control and adjust the execution of the grasping and / or predetermined operation actions.

9. The transformable mobility vehicle of claim 3, wherein, The structural state feedback adjustment component includes: a state monitoring unit configured to monitor the extension amount and / or stress of the rods based on the internal state information acquired by the perception component in real time; an adjustment controller configured to output a compensation control signal to adjust the structural configuration of the variable geometry truss to achieve stress redistribution or configuration error correction when the monitored data deviates from a preset safe range or target parameter.

10. The transformable mobility vehicle of claim 9, wherein, The adjustment controller includes: a storage module configured to pre-store the safe stroke range and / or safe stress range of each rod, and pre-set the allowable configuration error threshold of the actual configuration of the variable geometry truss relative to the target configuration; A control module is configured to read the telescopic and / or force conditions collected by the state monitoring unit. When it is detected that the current telescopic length or force of any rod exceeds the corresponding safety range, and / or the deviation between the actual geometric truss configuration determined based on the current lengths of all rods and the target configuration exceeds the allowed shape error threshold, it is determined that the preset safety range or target parameters are deviated, and a compensation control signal is generated according to the type and amplitude of the deviation to adjust the target telescopic length of one or more rods with active length change.

Images