CN-121989225-A - Whole-body partition self-adaptive thermal management system of intelligent robot

CN121989225ACN 121989225 ACN121989225 ACN 121989225ACN-121989225-A

Abstract

The invention discloses a whole body partition self-adaptive heat management system of an intelligent robot, which relates to the technical field of robot heat management and comprises a heat management control center, wherein the heat management control center is in communication connection with a visual perception identification module, and the visual perception identification module is used for identifying the type, action intensity and environment temperature of an operation object of the robot in real time through a visual system and judging heat management requirements in combination with task types. According to the invention, the physical properties of the operation object, the action intensity of the robot and the environmental temperature are analyzed in real time by integrating the visual recognition and the environmental perception technology, so that the thermal management requirements under different task scenes can be accurately judged, the thermal management resources can be distributed to each key part of the robot according to the requirements, the thermal state is always matched with the task requirements when the robot executes fine operation or high-intensity movement, and the overall performance and the task adaptability of the robot under diversified and complicated operation scenes are obviously improved.

Inventors

LIU BO
WU XIAOHUI
ZHANG XIANGPING
ZHENG JINJUN

Assignees

温擎智控(上海)机器人有限公司

Dates

Publication Date: 20260508
Application Date: 20251222

Claims (10)

1. The whole body partition self-adaptive heat management system of the intelligent robot comprises a heat management control center, and is characterized in that the heat management control center is in communication connection with the following modules: The visual perception recognition module is used for recognizing the type, action intensity and environment temperature of the robot operation object in real time through the visual system, and judging the thermal management requirement by combining the task type; the thermal demand analysis module is used for collecting the temperature and motion state data of each joint in real time, and carrying out thermal load analysis by combining the visual recognition result to form a partition thermal management instruction; The access switching control module is used for receiving the partition thermal management instruction, dynamically controlling the access connection mode of the multi-way valve, and carrying out mode switching of different access combinations by using the multi-way valve; the mode self-adaptive execution module is used for automatically executing a preset or dynamically generated thermal management mode according to the current task and the environment state; The partition thermal management execution module is used for executing specific thermal management actions, including cooperative control of a radiator, a fan, a water pump and a PTC heater, and carrying out thermal balance evaluation according to system energy efficiency and thermal load change so as to realize balance optimization of energy consumption and heat dissipation effect.
2. The system for self-adaptive thermal management of whole body partition of intelligent robot according to claim 1, wherein said visual perception recognition module comprises an object motion recognition unit and an ambient temperature perception unit; the object action recognition unit is used for recognizing the attribute of an operation object and the current action intensity of the robot by utilizing the visual sensor and judging whether the operation object belongs to a high-load fine task or not; The environment temperature sensing unit comprehensively detects external temperature conditions through the visual sensor and the environment sensor, recognizes whether a heating mode needs to be started or not, and determines a preheating strategy by combining a task type.
3. The adaptive thermal management system for whole body partition of intelligent robot according to claim 2, wherein the object motion recognition unit performs the steps of: Capturing image information of an operation object in real time through a visual sensor integrated on the head of the robot, and identifying physical properties of the operation object based on a pre-trained deep learning model, including whether the operation object is a fragile product, a weight grade and required operation precision; combining real-time current and position feedback data of the robot joint, comprehensively analyzing the motion speed, the load moment and the motion repetition frequency of the current mechanical arm or smart hand, and quantitatively evaluating the motion intensity of the current task; Based on the physical attribute of the identified operation object and the quantized action intensity, judging whether the current task belongs to a high-load fine task, and generating a corresponding upper limb joint heat dissipation priority instruction for a thermal management control center so as to allocate cooling resources in advance.
4. The system for self-adaptive thermal management of whole body partition of intelligent robot according to claim 2, wherein said ambient temperature sensing unit is executed as follows: detecting the environment temperature condition of the robot by fusing scene analysis of the vision sensor and reading of the environment temperature sensor, and identifying whether a low-temperature working condition exists; Determining a preheating strategy under a low-temperature working condition by combining the task type determined by the object action recognition unit, wherein the preheating strategy comprises the steps of judging whether a key motion joint for executing a current task needs to be heated preferentially or not; Based on the determined warming strategy, a warming command including the target joint and the heating path is generated and sent to the thermal management control center to initiate an environmentally adaptive heating cycle.
5. The system for self-adaptive thermal management of whole body partition of intelligent robot according to claim 2, wherein said thermal demand analysis module comprises a joint temperature monitoring unit and a thermal load analysis unit The joint temperature monitoring unit is used for monitoring the temperature change of each region in real time through temperature sensors arranged on the joints of the upper limbs, the lower limbs and the trunk so as to analyze the abnormal change trend of the temperature, identify the local overheat or low temperature region and generate a thermal state map; The thermal load analysis unit is used for analyzing thermal load distribution according to the joint movement data and abnormal temperature change trend, identifying high-load joints and predicting the thermal management requirements of the high-load joints.
6. The adaptive thermal management system for whole body partition of intelligent robot according to claim 5, wherein the joint temperature monitoring unit performs the steps of: synchronously acquiring real-time temperature data of each region at a preset sampling frequency through temperature sensor arrays embedded in the upper limb joint module, the lower limb joint module and the trunk controller; filtering and time sequence analysis are carried out on the acquired real-time temperature data, and the temperature rise rate and the temperature gradient of each joint in unit time are calculated so as to identify the abnormal change trend of the temperature; And positioning a local area in the overheat or supercooled state based on the abnormal temperature change trend, and generating a thermal state map containing area identifiers and temperature states.
7. The adaptive thermal management system for whole body partition of intelligent robot according to claim 5, wherein the thermal load analysis unit performs the steps of: receiving a thermal state map from a joint temperature monitoring unit, and synchronously acquiring real-time motion data of each joint motor, wherein the real-time motion data comprise power output, rotating speed and efficiency parameters; Calculating the heat generation rate and the heat dissipation capacity of each joint at present through a thermodynamic model based on the thermal state map and the real-time motion data, and drawing a whole-body thermal load distribution map; According to the whole body thermal load distribution diagram, the thermal management demand of the high-load joint in a short period in the future is predicted, and a predictive partition thermal management instruction is generated and uploaded to a thermal management control center, wherein the partition thermal management instruction comprises joint identification and regulation intensity which need to be subjected to heat dissipation or heating preferentially.
8. The adaptive thermal management system for whole body partition of intelligent robot according to claim 5, wherein said path switching control module performs the steps of: Receiving a partition thermal management instruction from a thermal management control center, and extracting a target heat dissipation or heating area and a required flow path configuration from the partition thermal management instruction; Analyzing corresponding multi-way valve passage combination logic according to the partition thermal management instruction, wherein the multi-way valve is a programmable control valve with at least eight interfaces; and an electromagnetic actuator for driving the multi-way valve is used for switching the valve core to a designated station, and establishing corresponding liquid cooling loop connection to realize directional circulation of cooling liquid among the trunk, the upper limb joint module, the lower limb joint module and the battery pack.
9. The system for self-adaptive thermal management of whole body partition of intelligent robot according to claim 8, wherein said mode self-adaptive execution module performs the steps of: Analyzing a mode execution instruction generated by the partition thermal management instruction, and monitoring the task state and the environmental parameters of the robot in real time; according to the current task and the environment state, a matched thermal management mode is called from a prestored mode library to generate a new flow path control sequence; and the coordination passage switching control module and the partition thermal management execution module execute the flow path control sequence in sequence and output corresponding action instructions.
10. The adaptive thermal management system for whole body partition of intelligent robot according to claim 9, wherein the execution steps of the partition thermal management execution module are as follows: Receiving an action instruction from a mode self-adaptive execution module, wherein the action instruction is used for defining target working states of a radiator, a fan, a water pump and a PTC heater; each actuator is independently or cooperatively controlled according to the action command, and the flow rate of the water pump, the rotating speed of the fan and the power of the PTC heater are regulated so as to execute specific heat dissipation or heating action; In the execution process, energy consumption data of the system and temperature feedback of key nodes are collected in real time, the thermal balance state is evaluated, the thermal balance coefficient is calculated, and the parameters of each actuator are dynamically adjusted in a fine mode.

Description

Whole-body partition self-adaptive thermal management system of intelligent robot Technical Field The invention relates to the technical field of robot thermal management, in particular to a whole body partition self-adaptive thermal management system of an intelligent robot. Background In the technical development process of the current intelligent robot with body, the movement and control performance of the robot are continuously improved, core components such as a joint motor, a main control unit and a battery generate a large amount of heat during working, and especially under a long-time or high-load operation scene, the heat accumulation problem is increasingly prominent, meanwhile, in a low-temperature environment, the phenomena of reduced battery activity, difficult starting of the components and the like also influence the normal operation capability of the robot, and in order to ensure the stability and reliability of the robot in a complex environment and various tasks, a thermal management mechanism capable of covering the whole body and regulating the temperature according to needs is needed, so that a heat dissipation or heating strategy is dynamically adjusted according to the actual working state and the external environment of each part of the robot, thereby realizing efficient and self-adaptive thermal management and supporting the continuous operation of the robot in a wider scene. When the robot executes fine operation tasks, such as assembling or grabbing fragile objects, the temperature of joints of the robot rises rapidly due to local continuous high load, the traditional heat dissipation mode cannot identify and respond to local overheating in real time, so that action delay or accuracy is reduced, and when the robot is in a low-temperature environment, the robot is in whole body preheating, if the robot only depends on a fixed heating path, the critical movement joints cannot be heated preferentially according to actual task requirements, so that the preheating efficiency is low, the response is slow, and the starting and movement performance of the robot in the low-temperature environment are affected. Disclosure of Invention In order to solve the technical problems, the technical scheme is that the whole body partition self-adaptive thermal management system of the intelligent robot comprises a thermal management control center, wherein the thermal management control center is in communication connection with the following modules: the visual perception recognition module is used for recognizing the type, action intensity and environment temperature of the robot operation object in real time through the visual system, judging the thermal management requirement by combining the task type, providing a dynamic input basis for a subsequent thermal control strategy, and realizing task-oriented thermal management prejudgment and response; The thermal demand analysis module is used for collecting the temperature and motion state data of each joint in real time, carrying out thermal load analysis by combining the visual recognition result, judging the joints needing to be preferentially radiated or heated, and forming a partition thermal management instruction; The channel switching control module is used for receiving the partition thermal management instruction, dynamically controlling the channel connection mode of the multi-way valve, and utilizing the multi-way valve to perform mode switching of different channel combinations, so as to realize flexible configuration of cooling and heating paths and improve the response speed and energy efficiency of the system; the mode self-adaptive execution module is used for automatically executing a preset or dynamically generated thermal management mode according to the current task and the environment state, guaranteeing the optimal execution of the thermal management of the robot under the complex working condition and enhancing the robustness of the system; The zoned thermal management execution module is used for executing specific thermal management actions, comprising cooperative control of a radiator, a fan, a water pump and a PTC heater, and carrying out thermal balance evaluation according to system energy efficiency and thermal load change so as to realize balance optimization of energy consumption and heat dissipation effect and ensure that the whole body thermal state of the robot is always in the optimal range required by a task. Preferably, the visual perception recognition module comprises an object action recognition unit and an environment temperature perception unit; The object action recognition unit is used for recognizing the attribute (such as fragile products) of an operation object and the current action intensity of the robot by utilizing the visual sensor, judging whether the operation object belongs to a high-load fine task or not, providing a basis for joint heat dissipation priority, avoiding action precision reduct