Lampblack robot with artificial intelligent lampblack discharge path planning function

Abstract

The invention provides a cooking fume robot with artificial intelligence for cooking fume discharge path planning, and relates to the technical field of cooking fume treatment.The cooking fume robot comprises a robot body module used for sucking and discharging cooking fume and internally provided with an air suction inlet and an air discharge outlet, and effective circulating treatment of the cooking fume is achieved; and the oil fume detection module is used for predicting the change trend of the oil fume concentration in a period of time in the future according to the historical oil fume concentration data and the current environmental parameters, and dynamically adjusting the oil fume treatment intensity through the self-adaptive regulation and control rate according to the predicted oil fume concentration change trend so as to filter oil fume in air flowing through the interior of the robot main body. By integrating the artificial intelligence technology, intelligent cooking fume treatment, dynamic path planning and autonomous obstacle avoidance are realized, and the cooking fume discharge efficiency and the flexibility and safety of kitchen operation are effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of oil fume treatment, and in particular to an oil fume robot with artificial intelligence and oil fume emission path planning. Background Art

[0002] Traditional fume exhaust systems often rely on fixed exhaust paths and preset operating modes. For example, in a restaurant kitchen, the exhaust ducts and fans may be arranged according to the original design. However, if the kitchen layout or cooking equipment changes, the original exhaust path may no longer be optimal.

[0003] For example, if a restaurant adds a large oven near the end of the existing exhaust path, this can lead to poor fume exhaust in that area, impacting the kitchen environment and employee health. In this case, traditional systems struggle to flexibly adjust the exhaust path to accommodate this change.

[0004] In addition, traditional oil fume exhaust equipment lacks the ability to autonomously avoid obstacles and adapt to environmental changes. In a complex kitchen environment, there may be obstacles such as temporarily stacked items, moving cooking equipment or people. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a fume robot with artificial intelligence oil fume emission path planning, which intelligently plans the oil fume emission path and realizes efficient and accurate oil fume emission.

[0006] In order to solve the above technical problems, the technical solutions of the present invention are as follows: In a first aspect, a fume robot with artificial intelligence-based fume emission path planning comprises: The robot's main module is used to inhale and exhaust cooking fumes, and has built-in air intake and exhaust ports to achieve effective recycling of cooking fumes. The oil fume detection module is used to predict the changing trend of oil fume concentration in the future based on historical oil fume concentration data and current environmental parameters. It dynamically adjusts the oil fume treatment intensity through adaptive control rate according to the predicted oil fume concentration trend to filter out oil fume in the air flowing through the robot body; The path planning module is used to automatically plan the final oil fume emission path based on the oil fume concentration distribution of the oil fume detection module, the location information of obstacles in the kitchen, and the preset kitchen layout using the frog sound search algorithm. It also controls the movement of the mobile unit according to the final oil fume emission path to achieve positioning and path tracking; The obstacle avoidance and circumvention module is used to detect obstacles around the robot in real time and control the robot's autonomous obstacle avoidance and circumvention operations based on the detection results; The control module is used to electrically connect with the oil fume detection module, the path planning module, and the obstacle avoidance and bypassing module to complete the oil fume emission task.

[0007] Furthermore, based on historical fume concentration data and current environmental parameters, the changing trend of fume concentration in the future is predicted. Based on the predicted trend of fume concentration, the fume treatment intensity is dynamically adjusted through an adaptive control rate to filter out fume from the air flowing through the robot body, including: The system uses sensors to obtain real-time environmental parameters, including temperature, humidity, and wind speed. It then extracts features related to changes in oil fume concentration from current environmental parameter data and historical oil fume concentration data, including historical trends in oil fume concentration and patterns of environmental parameter changes. Time series analysis is used as a trend prediction model. Historical oil fume concentration data and corresponding environmental parameters are used as training sets to train the trend prediction model and obtain a trained prediction model. Input the current environmental parameters into the trained prediction model to generate the prediction results of the change trend of oil smoke concentration in the future; Calculate the adaptive control rate at the current moment based on the predicted results of the change trend of oil smoke concentration in the future; A basic fume treatment intensity is set, and the fume treatment intensity is adjusted in real time according to the adaptive control rate to filter out the fume in the air flowing through the robot body.

[0008] Furthermore, based on the predicted results of the change trend of the oil smoke concentration in the future, the adaptive control rate at the current moment is calculated, including: The actual oil fume concentration measurement value at the current moment is obtained through the oil fume concentration sensor, and the predicted oil fume concentration is determined based on the current ambient temperature, humidity and historical oil fume concentration data; According to the predicted oil smoke concentration and the actual oil smoke concentration measurement value at the current moment, the change in oil smoke concentration is obtained; The change in oil fume concentration is compared with the preset maximum oil fume concentration and normalized to obtain a relative change rate; based on the relative change rate, the adaptive control rate at the current moment is determined.

[0009] Furthermore, based on the fume concentration distribution of the fume detection module, the location information of obstacles in the kitchen, and the preset kitchen layout, a frog sound search algorithm is used to automatically plan the final fume emission path, and the mobile unit is controlled to move according to the final fume emission path to achieve positioning and path tracking, including: Integrate the oil smoke concentration distribution data, obstacle location information and preset kitchen layout information to form a comprehensive information set; Based on the comprehensive information set, the frog sound search algorithm is initialized, including setting the number of frogs, the initial position of each frog, the number of iterations, and the key parameters of the frog sound search algorithm, including the local search range and the global search range; Each frog explores the path within its local search range by adjusting key points on the path, including the starting point, turning point, and end point, and the frog group shares search experience through global information exchange; The quality of each path is evaluated by calculating the path quality scoring function value, and the path quality scoring result is obtained. The position of the frog group is updated according to the path quality scoring result. The iterative process of local search and path adjustment, global information exchange and frog group update is repeated until the preset number of iterations is reached, and the final oil smoke emission path is determined from the frog group based on the frog's path quality score; The final planned oil fume emission path is converted into an instruction sequence understood by the mobile part, including moving direction, speed, and acceleration, and the mobile part is controlled to move according to the instruction sequence to achieve positioning and path tracking.

[0010] Furthermore, the key parameters of the frog sound search algorithm include a local search range and a global search range.

[0011] Furthermore, the quality of each path is evaluated by calculating the path quality scoring function value to obtain the path quality scoring result, and the position of the frog group is updated according to the path quality scoring result, including: Evaluate the average oil fume concentration at the measurement points along the route, and determine the weighted value of the oil fume concentration factor in the total score based on the average oil fume concentration; Measure the path length and obtain the weighted value of the path length factor in the total score; Calculate the average of the reciprocal of the distance from each point on the path to the nearest obstacle, and based on the average, evaluate the weighted contribution of the obstacle proximity factor in the total score; Analyze the smoothness between adjacent points on the path and integrate the smoothness of all adjacent points to obtain the weighted contribution of the path smoothness factor in the total score; The weighted value of the oil smoke concentration factor in the total score, the weighted value of the path length factor in the total score, the weighted contribution of the obstacle proximity factor in the total score, and the weighted contribution of the path smoothness factor in the total score are combined to obtain the total quality score of the path.

[0012] Furthermore, the robot can detect obstacles around it in real time and control the robot's autonomous obstacle avoidance and circumvention operations based on the detection results, including: Use lidar, ultrasonic sensors, and infrared sensors to detect obstacles in the surrounding environment in real time and identify the location, shape, and size of the obstacles; Compare obstacle information with pre-set kitchen layout information to determine the location and nature of the obstacle, and combine it with the robot's current position, speed, and direction of movement to assess the potential impact of the obstacle on the robot's motion path; Develop obstacle avoidance and circumvention strategies based on obstacle information and the robot's motion state, including adjusting the fume exhaust path, changing the movement direction, reducing speed, or accelerating to bypass obstacles; According to the obstacle avoidance and circumvention strategies, corresponding control instructions are generated, including movement direction, speed and acceleration, and the control instructions are transmitted to the mobile part of the robot. The mobile part performs corresponding motion operations according to the instructions to realize the obstacle avoidance and circumvention functions.

[0013] Furthermore, the obstacle avoidance and circumvention strategies include adjusting the fume emission path, changing the moving direction, reducing the speed or accelerating to circumvent obstacles.

[0014] In a second aspect, a computing device includes: one or more processors; The storage device is used to store one or more programs, and when the one or more programs are executed by the one or more processors, the one or more processors implement the method.

[0015] According to a third aspect, a computer-readable storage medium stores a program, which implements the method described above when executed by a processor.

[0016] The above solution of the present invention includes at least the following beneficial effects: The robot's main module features built-in air intakes and exhaust ports for efficient fume recirculation. This design ensures that fume is quickly drawn in and processed, reducing its residence time in the kitchen and improving fume treatment efficiency. The fume detection module predicts future trends in fume concentration based on historical fume concentration data and current environmental parameters. This predictive capability enables the robot to proactively adjust fume treatment intensity to accommodate changes in fume concentration, resulting in more efficient air filtration of fume.

[0017] The application of adaptive control rates enables the robot to dynamically adjust processing intensity based on actual conditions, avoiding over- or under-processing, saving energy, and improving processing effectiveness. The path planning module automatically plans the final fume exhaust path using a frog-like search algorithm. This algorithm boasts strong global search capabilities and rapid convergence, ensuring the robot finds the optimal or suboptimal exhaust path. By considering the distribution of fume concentration, the location of kitchen obstacles, and the preset kitchen layout, the path planning module generates an exhaust path that meets actual needs, improving the robot's intelligence and efficiency.

[0018] The obstacle avoidance and avoidance module can detect obstacles around the robot in real time and control the robot's autonomous obstacle avoidance and avoidance operations based on the detection results. This capability enables the robot to move flexibly in complex kitchen environments, avoid collisions with obstacles, and protect the safety of the robot and kitchen facilities.

[0019] The control module is electrically connected to the fume detection module, path planning module, and obstacle avoidance and avoidance module, enabling information sharing and collaborative operation among these modules. This integrated design enables the robot to complete fume removal tasks more efficiently, improving the performance and reliability of the overall system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 This is a schematic diagram of a fume robot with artificial intelligence fume emission path planning provided by an embodiment of the present invention.

[0021] Figure 2 This is a flowchart of an oil fume emission path planning system with artificial intelligence provided by an embodiment of the present invention, which automatically plans the final oil fume emission path based on the oil fume concentration distribution of the oil fume detection module, the location information of obstacles in the kitchen and the preset kitchen layout, and controls the movement of the mobile part according to the final oil fume emission path to achieve positioning and path tracking. DETAILED DESCRIPTION

[0022] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0023] like Figure 1 As shown, an embodiment of the present invention provides a fume robot with artificial intelligence-based fume exhaust path planning, comprising: The robot main body module 1 is used to inhale and exhaust oil smoke, and has built-in air intake and air exhaust ports to achieve effective circulation and treatment of oil smoke; Fume detection module 2 is used to predict the changing trend of fume concentration in the future based on historical fume concentration data and current environmental parameters, and dynamically adjust the fume treatment intensity through adaptive control rate according to the predicted fume concentration changing trend to filter out fume in the air flowing through the robot body; Path planning module 3 is used to automatically plan the final oil fume emission path using a frog sound search algorithm based on the oil fume concentration distribution of the oil fume detection module, the location information of obstacles in the kitchen, and the preset kitchen layout, and control the movement of the mobile unit according to the final oil fume emission path to achieve positioning and path tracking; Obstacle avoidance and circumvention module 4, used to detect obstacles around the robot in real time, and control the robot's autonomous obstacle avoidance and circumvention operations based on the detection results; The control module 5 is used to be electrically connected to the oil fume detection module, the path planning module, and the obstacle avoidance and bypassing module to complete the oil fume emission task.

[0024] In this embodiment of the present invention, the built-in air intake and exhaust ports enable efficient intake and exhaust of cooking fumes, effectively circulating and treating them and maintaining fresh kitchen air. The cooking fume detection module 2 accurately predicts future trends in cooking fume concentration based on historical cooking fume concentration data and current environmental parameters, providing a scientific basis for cooking fume treatment. Dynamically adjusting the cooking fume treatment intensity through an adaptive control rate ensures optimal cooking fume removal, avoiding energy waste caused by overtreatment while ensuring thorough cooking fume treatment.

[0025] Path Planning Module 3 uses a frog-like search algorithm to automatically plan the fume exhaust path, improving its efficiency and accuracy, enabling the robot to quickly find the final exhaust path. By combining information about fume concentration distribution and the locations of kitchen obstacles, it achieves precise positioning and path tracking, ensuring smooth and efficient fume exhaust.

[0026] The obstacle avoidance and circumvention module 4 detects obstacles around the robot in real time, effectively preventing collisions and protecting the robot's safety. Based on the detection results, it autonomously performs obstacle avoidance and circumvention operations, enhancing the robot's autonomy and intelligence, enabling it to maneuver flexibly in complex kitchen environments.

[0027] The control module 5 serves as the "brain" of the entire robot. It electrically connects to the fume detection module, path planning module, and obstacle avoidance and avoidance module, coordinating the work of each module to ensure the successful completion of the fume emission task. This integrated control design simplifies the robot's operation, improving ease of use and efficiency.

[0028] In a preferred embodiment of the present invention, based on historical oil fume concentration data and current environmental parameters, a change trend of oil fume concentration in a future period of time is predicted, and the oil fume treatment intensity is dynamically adjusted through an adaptive control rate according to the predicted change trend of oil fume concentration to filter out oil fume in the air flowing through the interior of the robot body. The method may include: The system uses sensors to obtain real-time environmental parameters, including temperature, humidity, and wind speed. It then extracts features related to changes in oil fume concentration from current environmental parameter data and historical oil fume concentration data, including historical trends in oil fume concentration and patterns of environmental parameter changes. Time series analysis is used as a trend prediction model. Historical oil fume concentration data and corresponding environmental parameters are used as training sets to train the trend prediction model and obtain a trained prediction model. Input the current environmental parameters into the trained prediction model to generate the prediction results of the change trend of oil smoke concentration in the future; Calculate the adaptive control rate at the current moment based on the predicted results of the change trend of oil smoke concentration in the future; A basic fume treatment intensity is set, and the fume treatment intensity is adjusted in real time according to the adaptive control rate to filter out the fume in the air flowing through the robot body.

[0029] In this embodiment of the present invention, after the machine is started, the internal sensor module begins operating, monitoring the temperature, humidity, and wind speed in the kitchen environment in real time. The sensor converts the collected data into electrical signals and transmits them to the data processing unit for further analysis. The data processing unit receives the environmental parameter data from the sensor and retrieves historical oil fume concentration data from the storage unit. Using a data analysis algorithm, it extracts features closely related to changes in oil fume concentration, such as historical trends in oil fume concentration (increasing, decreasing, or stable) and patterns in environmental parameter changes (such as the correlation between temperature and humidity and the impact of wind speed on oil fume diffusion). Time series analysis is selected as the trend prediction model because changes in oil fume concentration often exhibit time series characteristics. The extracted feature data (historical oil fume concentration and corresponding environmental parameters) is used as a training set and input into the time series analysis model. The model is trained using a model training algorithm, and its parameters are adjusted to accurately predict future trends in oil fume concentration. After training is complete, the real-time environmental parameter data is input into the trained prediction model. The prediction model, based on the input environmental parameters and combined with previously trained parameters, generates a prediction of oil fume concentration trends over a period of time.

[0030] The data analysis unit receives the prediction results of the oil fume concentration trend generated by the prediction model. Based on the prediction results and the preset control strategy (such as increasing the processing intensity when the oil fume concentration rises and reducing the processing intensity when it falls), the adaptive control rate at the current moment is calculated. When designing the robot, a basic oil fume treatment intensity is set as the default value. When the robot is working, the oil fume treatment intensity is adjusted in real time based on the calculated adaptive control rate. If the prediction results show that the oil fume concentration will rise, the treatment intensity is increased; if the prediction results show that the oil fume concentration will fall, the treatment intensity is reduced. By adjusting the oil fume treatment intensity, it is ensured that the oil fume in the air flowing through the interior of the robot body can be effectively filtered out.

[0031] Suppose a robot is operating in a kitchen with a current temperature of 25°C, humidity of 60%, and wind speed of 0.5 m / s. The robot acquires these environmental parameters in real time through sensors and retrieves historical oil fume concentration data from its storage unit. Using a data analysis algorithm, the robot discovers that oil fume concentration tends to increase when temperature rises, humidity decreases, and wind speed decreases. Therefore, the robot uses time series analysis as a trend prediction model and trains it using historical data. After training, the robot inputs the current environmental parameters into the prediction model, generating a forecast of oil fume concentration trends over the next period of time. The prediction indicates that oil fume concentration will gradually increase over the next 10 minutes. Based on this prediction, the robot calculates the adaptive control rate at the current moment and decides to increase the intensity of oil fume treatment. The robot then adjusts the parameters of the oil fume treatment module to improve treatment efficiency and ensure effective filtration of oil fume from the air flowing through the robot body.

[0032] By acquiring real-time environmental parameters and oil fume concentration data and combining it with a time series analysis model for prediction, the robot can proactively adjust its oil fume treatment intensity to ensure timely and effective filtration of oil fumes, thereby improving oil fume treatment efficiency. By adaptively adjusting the oil fume treatment intensity, the robot avoids unnecessary energy waste. When oil fume concentration is low, the robot automatically reduces the treatment intensity; when it rises, it increases the intensity. This intelligent control approach helps save energy costs. By adjusting the oil fume treatment intensity in real time, the robot ensures that oil fumes are effectively removed from the air flowing through its body, maintaining a fresh and clean kitchen air quality, which contributes to a better working environment and employee health. By integrating sensors, data analysis, and time series analysis models, the robot achieves intelligent prediction and adaptive control of oil fume concentration. This makes the robot more intelligent and autonomous, enabling it to better adapt to different kitchen environments and oil fume treatment needs.

[0033] In another preferred embodiment of the present invention, calculating the adaptive control rate at the current moment based on the prediction result of the change trend of the oil smoke concentration in the future period may include: The actual oil fume concentration measurement value at the current moment is obtained through the oil fume concentration sensor, and the predicted oil fume concentration is determined based on the current ambient temperature, humidity and historical oil fume concentration data; According to the predicted oil smoke concentration and the actual oil smoke concentration measurement value at the current moment, the change in oil smoke concentration is obtained; The change in oil fume concentration is compared with the preset maximum oil fume concentration and normalized to obtain a relative change rate; based on the relative change rate, the adaptive control rate at the current moment is determined.

[0034] In the embodiment of the present invention, the actual oil smoke concentration measurement value at the current moment is obtained from the oil smoke concentration sensor. , the unit is . Get historical oil smoke concentration data (past The concentration value at each time point) in units of . Get the current ambient temperature from the temperature sensor . Get the current ambient humidity from the humidity sensor The unit is percentage (%). The maximum value of preset oil smoke concentration , the unit is , used to normalize the change in oil smoke concentration.

[0035] In order to ensure that all parameters are at the same level, temperature and humidity need to be normalized. The minimum-maximum normalization method is used to map temperature and humidity to interval.

[0036] Temperature normalization: Assume the temperature range is , the normalized temperature .

[0037] Humidity normalization: Assume that the humidity range is , normalized humidity .

[0038] Select an appropriate time series analysis model (such as linear regression). Use historical oil smoke concentration data , normalized temperature and humidity As a feature variable. Train the model to predict the oil smoke concentration in the future ,here are the parameters obtained from model training.

[0039] Calculate the predicted oil smoke concentration The actual oil smoke concentration The difference between the two, that is, the change in oil smoke concentration .

[0040] Normalize the changes: The change in oil smoke concentration The maximum value of the preset oil smoke concentration Compare and get the relative rate of change .

[0041] Set two parameters and , used to adjust the range and benchmark value of the adaptive control rate. Assume =0.3 (indicates that the maximum range of the control rate that can change with the relative rate of change is 30%) and =0.5 (indicates that the control rate is 50% in steady state). According to the relative rate of change, the adaptive control rate at the current moment is calculated using the formula The calculated adaptive control rate Applied to the oil fume treatment module to dynamically adjust the treatment intensity.

[0042] Normalization ensures dimensional consistency among the parameters in the formula. By real-time monitoring of oil fume concentration and environmental parameters, and combining it with a predictive model to calculate an adaptive control rate, the robot can precisely control the intensity of the oil fume treatment module to ensure that oil fume concentration remains within the ideal range. The adaptive control rate dynamically adjusts the treatment intensity based on changes in oil fume concentration, avoiding over- or under-treatment and thus improving the efficiency of oil fume treatment. By precisely controlling the intensity of oil fume treatment, the robot avoids unnecessary waste of energy and resources and reduces operating costs. By continuously monitoring and dynamically adjusting the intensity of oil fume treatment, the robot ensures that kitchen air remains fresh, improving the working environment and the health of employees. This formula combines sensor data, historical data, and predictive models to achieve intelligent prediction and adaptive control of oil fume concentration, enhancing the robot's intelligence and adaptability.

[0043] In a preferred embodiment of the present invention, based on the fume concentration distribution of the fume detection module, the location information of obstacles in the kitchen, and the preset kitchen layout, a frog sound search algorithm is used to automatically plan the final fume emission path, and the mobile unit is controlled to move according to the final fume emission path to achieve positioning and path tracking, which may include: Integrate the oil smoke concentration distribution data, obstacle location information and preset kitchen layout information to form a comprehensive information set; Based on the comprehensive information set, the frog sound search algorithm is initialized, including setting the number of frogs, the initial position of each frog, the number of iterations, and the key parameters of the frog sound search algorithm, including the local search range and the global search range; Each frog explores the path within its local search range by adjusting key points on the path, including the starting point, turning point, and end point, and the frog group shares search experience through global information exchange; The quality of each path is evaluated by calculating the path quality scoring function value, and the path quality scoring result is obtained. The position of the frog group is updated according to the path quality scoring result. The iterative process of local search and path adjustment, global information exchange and frog group update is repeated until the preset number of iterations is reached, and the final oil smoke emission path is determined from the frog group based on the frog's path quality score; The final planned oil fume emission path is converted into an instruction sequence understood by the mobile part, including moving direction, speed, and acceleration, and the mobile part is controlled to move according to the instruction sequence to achieve positioning and path tracking.

[0044] In this embodiment of the present invention, the fume detection module acquires real-time fume concentration distribution data. This data is represented as a grid or point cloud, with each point containing the fume concentration value at that location. Simultaneously, various sensors, such as lidar and cameras, are used to acquire the location information of obstacles within the kitchen, including their type, size, and coordinates. Pre-set kitchen layout information, such as the location and dimensions of walls, doors, windows, and kitchen equipment, is also acquired. The fume concentration distribution data, obstacle location information, and pre-set kitchen layout information are integrated to form a comprehensive information set containing all necessary information. This information set serves as input to the frog sound search algorithm.

[0045] The number of frog swarms is determined based on the size and complexity of the kitchen. A larger swarm increases the search range, but also increases the computational effort. Each frog's initial position is determined randomly or based on a strategy (such as the oil smoke concentration gradient). The number of iterations is set based on actual needs to ensure a thorough search process.

[0046] The local and global search ranges are set based on the size of the kitchen and the planning requirements for the fume emission path. The frog search algorithm is initialized based on the set parameters. Within the local search range, each frog explores the path by adjusting key points along the path (starting point, turning point, and end point). During the exploration process, the frog considers factors such as fume concentration and obstacle avoidance rate to find an optimal path. During the exploration process, the frog adjusts the path based on a path quality scoring function to improve path quality. This path quality scoring function comprehensively considers multiple factors, including fume concentration, path length, and obstacle avoidance rate.

[0047] Frog groups share their search experiences through some information exchange mechanism (such as broadcasting, point-to-point communication).

[0048] Each frog can receive the path quality score results and search strategies of other frogs, thereby adjusting its own search direction. According to the preset path quality score function, the quality of each path is evaluated, and the path quality score results will serve as the basis for updating the frog group. The position and status information of the frog group are updated based on the path quality score results. The iterative process of local search and path adjustment, global information exchange and frog group update is repeated continuously. After each iteration, the position and status information of the frog group, as well as the path quality score results, are updated. When the preset number of iterations is reached, the machine determines the final oil fume emission path from the frog group based on the path quality score of the frog. The final planned oil fume emission path is converted into a command sequence that can be understood by the mobile unit, including movement direction, speed, acceleration, etc. The generation of the command sequence needs to take into account the motion characteristics and control accuracy of the mobile unit.

[0049] The mobile part is controlled to move according to the instruction sequence to achieve positioning and path tracking. During the movement process, the machine will monitor the position and status information of the mobile part in real time to ensure that the mobile part moves according to the planned path.

[0050] Assuming that the fume concentration in the kitchen is unevenly distributed and there are multiple obstacles such as cabinets and stoves, the frog sound search algorithm is used to plan the fume emission path.

[0051] The frog swarm was initialized with 20 individuals, initially randomly distributed within the kitchen. The number of iterations was set to 500. The local search range was set to 10 grid points around the frog's current location, and the global search range was set to the entire kitchen. Each frog adjusted key points on its path within the local search range, attempting to find a path with low smoke concentration and high obstacle avoidance. The frog swarm shared its search experience via broadcasting, and each frog adjusted its search strategy based on the path quality scores received from other frogs.

[0052] After each iteration, the machine evaluates the quality of each path and updates the frog population's position and status based on the path quality scores. Frogs with high-quality paths are retained, while those with low-quality paths are eliminated, or frogs are relocated or mutated. After 500 iterations, the machine selects the frog with the highest path quality from the population and determines the final fume emission path. The final planned fume emission path is converted into a sequence of instructions that the mobile unit can understand, and the mobile unit is controlled to move according to the planned path.

[0053] The frog-sound search algorithm automatically plans a short exhaust path with lower fume concentration, thereby improving fume emission efficiency. The planned exhaust path avoids areas with high fume concentration, effectively reducing fume concentration in the kitchen and improving the kitchen environment. By converting the planned exhaust path into a sequence of instructions understandable by the mobile unit and controlling it to move along the planned path, the mobile unit's path-tracking capability is enhanced. The frog-sound search algorithm can adapt to different kitchen layouts and fume concentration distributions, demonstrating strong adaptability and robustness. The entire process requires no human intervention, enhancing the automation level of exhaust path planning and reducing labor costs.

[0054] In another preferred embodiment of the present invention, key parameters of the frog sound search algorithm include a local search range and a global search range.

[0055] In this embodiment of the present invention, a detailed analysis of the kitchen's layout, dimensions, and fume concentration distribution is performed. This includes identifying obstacles, wall orientation, and door and window locations within the kitchen, as well as acquiring real-time fume concentration data through a fume detection module. Based on this analysis of the kitchen environment, the robot determines the granularity of its local search. This granularity, which is related to the size of the grid or the density of the point cloud, determines the level of detail the frog can explore within its local search range.

[0056] Taking into account the search granularity and the actual size of the kitchen, the robot sets a local search range for each individual frog. This range is a rectangular or circular area centered on the frog's current position, large enough to allow the frog to explore the surrounding fume concentration changes and obstacle distribution. During the search process, the robot can dynamically adjust the size of the local search range based on the frog's search progress and real-time changes in fume concentration to improve search efficiency. The global search range needs to cover the entire kitchen area to ensure that the frog group can explore all possible paths. This requires the robot to have complete kitchen layout information and fume concentration distribution data.

[0057] Based on global information about the kitchen, the robot sets global search boundaries. This is achieved by defining the kitchen's four boundaries (up, down, left, right, or front, back, left, and right) to ensure the frog swarm stays within the kitchen during the search. This global search range not only defines the area the frog swarm can explore but also facilitates information exchange among the swarm. Through global broadcasts or point-to-point communication, the swarm can share search experience, improving overall search efficiency. If the kitchen environment changes (such as the movement of obstacles or a sudden increase in smoke concentration), the robot needs to adjust the global search range accordingly to ensure the swarm can continue to effectively explore the path.

[0058] In another preferred embodiment of the present invention, the quality of each path is evaluated by calculating a path quality scoring function value to obtain a path quality scoring result, and the position of the frog group is updated according to the path quality scoring result, including: Evaluate the average oil fume concentration at the measurement points along the route, and determine the weighted value of the oil fume concentration factor in the total score based on the average oil fume concentration; Measure the path length and obtain the weighted value of the path length factor in the total score; Calculate the average of the reciprocal of the distance from each point on the path to the nearest obstacle, and based on the average, evaluate the weighted contribution of the obstacle proximity factor in the total score; Analyze the smoothness between adjacent points on the path and integrate the smoothness of all adjacent points to obtain the weighted contribution of the path smoothness factor in the total score; The weighted value of the oil smoke concentration factor in the total score, the weighted value of the path length factor in the total score, the weighted contribution of the obstacle proximity factor in the total score, and the weighted contribution of the path smoothness factor in the total score are combined to obtain the total quality score of the path.

[0059] In the embodiment of the present invention, setting The oil fume concentration sensor is used to measure at each measuring point to obtain the oil fume concentration value. (The unit is mg / m³, which indicates the mass of oil smoke particles per cubic meter of air). Use the formula Calculate the average oil smoke concentration at the measurement points on the path, where is the concentration adjustment coefficient, which is used to adjust the impact of oil smoke concentration. It can be 1 (i.e. use the concentration value directly), greater than 1 (emphasize the effect of high concentration), or less than 1 (reduce the effect of low concentration). Substitute the average oil smoke concentration into the formula , get the weighted value of the fume concentration factor in the total score, where, It is the weight of the oil smoke concentration factor and is set according to actual needs.

[0060] Measure the total length of the path using a measuring tool (such as a tape measure, laser distance meter, etc.) (Units can be meters). Substituting the path length into the formula , get the weighted value of the path length factor in the total score, where, Is the weight of the path length factor, set according to actual needs. For each point on the path, use a measuring tool or sensor to calculate the distance to the nearest obstacle. (Units can be meters). Use the formula To calculate the average of the reciprocal distance from each point on the path to the nearest obstacle, It is the distance adjustment coefficient, which is used to adjust the impact of the obstacle proximity. Appropriate values can be used to emphasize or weaken the impact of the proximity of obstacles. Substitute the average value of the reciprocal of the distance into the formula , we get the weighted contribution of the obstacle proximity factor in the total score, where It is the weight of the obstacle proximity factor and is set according to actual needs.

[0061] For adjacent points on the path and , calculate the transition smoothing value between them The conversion smoothing value is defined based on factors such as the angle change and distance change between adjacent points. Specifically, one of the following methods can be used: Vector angle: Calculate from point Arrive The vector from point Arrive The angle between the vectors of , the smaller the angle, the smoother the path direction change. Sum the smoothness of all adjacent points to get the weighted contribution of the path smoothness factor in the total score, that is, ,in, The weight of the conversion smoothing factor is set according to actual needs. Substitute the weighted value of the fume concentration factor, the weighted value of the path length factor, the weighted contribution of the obstacle proximity factor and the weighted contribution of the conversion smoothing factor into the total quality score formula to obtain the total quality score of the path. , which is used to evaluate the overall quality of the path, .

[0062] This formula comprehensively considers multiple factors such as oil smoke concentration, path length, obstacle proximity, and transition smoothness, and can more comprehensively evaluate the quality of the path. By quantifying the contribution of each factor to the overall score, this formula provides a clear optimization goal for the path planning algorithm, helping the algorithm to find the optimal path more efficiently. The weighting coefficient in the formula This formula can be adjusted based on actual conditions, ranging from a constant 0.25 to other values, to adapt the algorithm to different kitchen environments and fume emission conditions. By optimizing the path quality score, this formula helps improve the range hood robot's path planning capabilities, enabling more efficient fume emission tasks and enhancing the user experience. This formula provides new insights and directions for path planning algorithm research, helping to promote technological advancement and development in related fields.

[0063] In a preferred embodiment of the present invention, real-time detection of obstacles around the robot and control of the robot's autonomous obstacle avoidance and circumvention operations based on the detection results may include: Use lidar, ultrasonic sensors, and infrared sensors to detect obstacles in the surrounding environment in real time and identify the location, shape, and size of the obstacles; Compare obstacle information with pre-set kitchen layout information to determine the location and nature of the obstacle, and combine it with the robot's current position, speed, and direction of movement to assess the potential impact of the obstacle on the robot's motion path; Develop obstacle avoidance and circumvention strategies based on obstacle information and the robot's motion state, including adjusting the fume exhaust path, changing the movement direction, reducing speed, or accelerating to bypass obstacles; According to the obstacle avoidance and circumvention strategies, corresponding control instructions are generated, including movement direction, speed and acceleration, and the control instructions are transmitted to the mobile part of the robot. The mobile part performs corresponding motion operations according to the instructions to realize the obstacle avoidance and circumvention functions.

[0064] In an embodiment of the present invention, the robot is equipped with a lidar sensor that emits a laser beam and receives the reflected signal. By analyzing the reflection time and angle of the laser beam, the distance and direction of the obstacle are calculated. The lidar continuously scans the surrounding environment, forming a 3D point cloud map, and updating the obstacle location information in real time. Multiple ultrasonic sensors are arranged on the surface of the robot, which emit ultrasonic waves and receive echoes. The distance to the obstacle is calculated based on the propagation time and speed of the ultrasonic waves. The ultrasonic sensor is sensitive to close-range obstacle detection and complements the blind spot of the lidar. The infrared sensor emits infrared light and receives the reflected light signal. By analyzing the intensity and pattern of the reflected light, the shape and size of the obstacle are identified. The infrared sensor is sensitive to heat sources and helps to identify heat-generating obstacles.

[0065] The robot obtains a kitchen layout diagram via the internet, including the location and dimensions of fixed obstacles (such as cabinets and stoves). This diagram is stored in coordinate format for easy comparison with real-time detection results. Obstacle information detected by lidar, ultrasonic, and infrared sensors is compared with the kitchen layout diagram to identify fixed obstacles or temporary obstacles (such as people or moving objects). The robot's current position, speed, and direction of movement are used to assess the potential impact of obstacles on its motion path. The robot's position, shape, size, and nature (fixed or temporary) are used to assess the degree of obstruction to the robot's motion path. The robot's motion state (such as speed and acceleration) is also considered to predict the impact of obstacles on its motion. If an obstacle blocks the intended fume exhaust path, the robot adjusts the exhaust path to bypass it. The robot changes its direction to select a path with fewer obstacles. Depending on the distance and nature of the obstacle, the robot reduces its speed to safely pass or accelerates to bypass it.

[0066] Based on the obstacle avoidance and avoidance strategies, the system generates corresponding control instructions, including movement direction, speed, and acceleration, ensuring that the instructions comply with the robot's motion capabilities and safety specifications. These control instructions are then transmitted to the robot's mobile components (e.g., motor controllers and drivers). The mobile components perform the corresponding motion operations, such as steering, acceleration, and deceleration, according to the instructions. The robot then adjusts its motion state in real time to maintain the effectiveness of its obstacle avoidance and avoidance functions.

[0067] Suppose that in a kitchen, a robot is performing a fume exhaust task and suddenly detects a moving person (temporary obstacle) in front of it.

[0068] The LiDAR detects a person ahead at a distance of approximately 2 meters. The ultrasonic sensor confirms the distance and detects the person's general outline. The infrared sensor senses the person's body temperature, further confirming that it is a moving person. The robot compares the detection results with the kitchen layout to confirm that there are no fixed obstacles at the location. Based on the robot's current position and speed, it assesses the potential impact of the person on the robot's motion path. The robot determines that the person would block the original fume exhaust path and decides to adjust its path, bypassing the person and reducing speed to ensure safety. The robot generates control commands for turning and deceleration and transmits them to the mobile unit, which then turns and decelerates to bypass the person. The robot continues its fume exhaust mission, maintaining the effectiveness of its obstacle avoidance function.

[0069] Real-time obstacle detection and autonomous obstacle avoidance effectively prevent collisions between the robot and obstacles, improving kitchen safety. The robot can adjust its motion path and speed in real time based on environmental changes, enhancing its flexibility and adaptability in complex kitchen environments. By optimizing its motion path and speed, the robot can more efficiently complete tasks such as fume removal and object handling. Autonomous obstacle avoidance and circumvention reduce the need for manual monitoring and intervention, reducing labor costs and workload, and enabling intelligent motion control for the robot, providing strong support for the development of kitchen automation and intelligence.

[0070] In another preferred embodiment of the present invention, the obstacle avoidance and circumvention strategy includes adjusting the fume emission path, changing the moving direction, reducing the speed or accelerating to circumvent obstacles.

[0071] In this embodiment of the present invention, a robot is in a kitchen working environment, with its various sensors (lidar, ultrasonic, and infrared) in operation. These sensors continuously collect real-time data about the surrounding environment, including the location, shape, size, and motion of obstacles. The robot's built-in algorithm processes and analyzes this sensor data to identify obstacles in the environment. Based on the location and nature of the obstacle (fixed or mobile), the robot assesses its potential impact on the robot's motion path. If an obstacle blocks the intended fume exhaust path, the robot replans the exhaust path. Using the built-in algorithm and kitchen layout information, the robot calculates a fume exhaust path that circumvents the obstacle. This new exhaust path is sent to the fume exhaust system for path adjustment. Based on the obstacle's location and the robot's current position, a path is calculated to avoid the obstacle. A steering command is generated to adjust the robot's movement direction, directing it along the new path. The robot executes the steering command and begins to change its direction. If the obstacle is close to the robot or its motion is unstable, the robot slows down and generates a deceleration command to adjust its speed to a safe range. The robot executes the deceleration command, reducing its speed to ensure safe passage through the obstacle area. When the robot is outside the obstacle zone and the path is clear, it accelerates and generates an acceleration command to increase its speed. The robot executes the acceleration command, quickly bypasses the obstacle, and continues to perform its task.

[0072] Based on the obstacle avoidance and avoidance strategies, the robot generates corresponding control instructions, which include parameters such as movement direction, speed, and acceleration. The control instructions are transmitted to the robot's mobile unit (such as the motor controller and driver). After receiving the control instructions, the mobile unit begins to execute the corresponding motion operations. According to the instructions, the motor speed, direction, and other parameters are adjusted to achieve the robot's movement, steering, acceleration, or deceleration. The robot moves according to the obstacle avoidance and avoidance strategies to ensure the safe and efficient completion of the task. During the execution process, the robot continuously monitors environmental changes and obstacle status. If a new obstacle is detected or the obstacle status changes, the robot will adjust the obstacle avoidance and avoidance strategies in real time. Based on real-time feedback, new control instructions are generated and transmitted to the mobile unit for execution.

[0073] An embodiment of the present invention further provides a computing device comprising: a processor and a memory storing a computer program, wherein the computer program, when executed by the processor, implements the operation of the robot described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0074] An embodiment of the present invention further provides a computer-readable storage medium storing instructions that, when executed on a computer, cause the computer to implement the aforementioned robot operation. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effects.

[0075] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A fume robot with artificial intelligence fume emission path planning, characterized in that: include: The robot's main module is used to inhale and exhaust cooking fumes, and has built-in air intake and exhaust ports to achieve effective recycling of cooking fumes. The oil fume detection module is used to predict the changing trend of oil fume concentration in the future based on historical oil fume concentration data and current environmental parameters. It dynamically adjusts the oil fume treatment intensity through adaptive control rate according to the predicted oil fume concentration trend to filter out oil fume in the air flowing through the robot body; The path planning module is used to automatically plan the final oil fume emission path based on the oil fume concentration distribution of the oil fume detection module, the location information of obstacles in the kitchen, and the preset kitchen layout using the frog sound search algorithm. It also controls the movement of the mobile unit according to the final oil fume emission path to achieve positioning and path tracking; The obstacle avoidance and circumvention module is used to detect obstacles around the robot in real time and control the robot's autonomous obstacle avoidance and circumvention operations based on the detection results; The control module is used to electrically connect with the oil fume detection module, the path planning module, and the obstacle avoidance and bypassing module to complete the oil fume emission task.

2. The oil fume robot with artificial intelligence oil fume exhaust path planning according to claim 1 is characterized in that: Based on historical fume concentration data and current environmental parameters, the changing trend of fume concentration in the future is predicted. The fume treatment intensity is dynamically adjusted through adaptive control rate based on the predicted fume concentration trend to filter out fume from the air flowing through the robot body, including: The system uses sensors to obtain real-time environmental parameters, including temperature, humidity, and wind speed. It then extracts features related to changes in oil fume concentration from current environmental parameter data and historical oil fume concentration data, including historical trends in oil fume concentration and patterns of environmental parameter changes. Time series analysis is used as a trend prediction model. Historical oil fume concentration data and corresponding environmental parameters are used as training sets to train the trend prediction model and obtain a trained prediction model. Input the current environmental parameters into the trained prediction model to generate the prediction results of the change trend of oil smoke concentration in the future; Calculate the adaptive control rate at the current moment based on the predicted results of the change trend of oil smoke concentration in the future; A basic fume treatment intensity is set, and the fume treatment intensity is adjusted in real time according to the adaptive control rate to filter out the fume in the air flowing through the robot body.

3. The oil fume robot with artificial intelligence oil fume exhaust path planning according to claim 2 is characterized in that: Based on the predicted results of the change trend of oil smoke concentration in the future, the adaptive control rate at the current moment is calculated, including: The actual oil fume concentration measurement value at the current moment is obtained through the oil fume concentration sensor, and the predicted oil fume concentration is determined based on the current ambient temperature, humidity and historical oil fume concentration data; According to the predicted oil smoke concentration and the actual oil smoke concentration measurement value at the current moment, the change in oil smoke concentration is obtained; The change in oil fume concentration is compared with the preset maximum oil fume concentration and normalized to obtain a relative change rate; based on the relative change rate, the adaptive control rate at the current moment is determined.

4. The oil fume robot with artificial intelligence oil fume exhaust path planning according to claim 3 is characterized in that: Based on the fume concentration distribution detected by the fume detection module, the location of obstacles in the kitchen, and the preset kitchen layout, the system automatically plans the final fume emission path using a frog sound search algorithm. The system then controls the mobile unit to move according to the final fume emission path, achieving positioning and path tracking. This includes: Integrate the oil smoke concentration distribution data, obstacle location information and preset kitchen layout information to form a comprehensive information set; Based on the comprehensive information set, the frog sound search algorithm is initialized, including setting the number of frogs, the initial position of each frog, the number of iterations, and the key parameters of the frog sound search algorithm, including the local search range and the global search range; Each frog explores the path within its local search range by adjusting key points on the path, including the starting point, turning point, and end point, and the frog group shares search experience through global information exchange; The quality of each path is evaluated by calculating the path quality scoring function value, and the path quality scoring result is obtained. The position of the frog group is updated according to the path quality scoring result. The iterative process of local search and path adjustment, global information exchange and frog group update is repeated until the preset number of iterations is reached, and the final oil smoke emission path is determined from the frog group based on the frog's path quality score; The final planned oil fume emission path is converted into an instruction sequence understood by the mobile part, including moving direction, speed, and acceleration, and the mobile part is controlled to move according to the instruction sequence to achieve positioning and path tracking.

5. The oil fume robot with artificial intelligence oil fume exhaust path planning according to claim 4 is characterized in that: The key parameters of the frog sound search algorithm include local search range and global search range.

6. The oil fume robot with artificial intelligence oil fume exhaust path planning according to claim 5 is characterized in that: The quality of each path is evaluated by calculating the path quality score function value, and the path quality score result is obtained. The position of the frog group is updated according to the path quality score result, including: Evaluate the average oil fume concentration at the measurement points along the route, and determine the weighted value of the oil fume concentration factor in the total score based on the average oil fume concentration; Measure the path length and obtain the weighted value of the path length factor in the total score; Calculate the average of the inverse of the distance from each point on the path to the nearest obstacle, and based on the average, evaluate the weighted contribution of the obstacle proximity factor in the total score; Analyze the smoothness between adjacent points on the path and integrate the smoothness of all adjacent points to obtain the weighted contribution of the path smoothness factor in the total score; The weighted value of the oil smoke concentration factor in the total score, the weighted value of the path length factor in the total score, the weighted contribution of the obstacle proximity factor in the total score, and the weighted contribution of the path smoothness factor in the total score are combined to obtain the total quality score of the path.

7. The oil fume robot with artificial intelligence oil fume exhaust path planning according to claim 6 is characterized in that: Detect obstacles around the robot in real time and control the robot's autonomous obstacle avoidance and circumvention operations based on the detection results, including: Use lidar, ultrasonic sensors, and infrared sensors to detect obstacles in the surrounding environment in real time and identify the location, shape, and size of the obstacles; Compare obstacle information with pre-set kitchen layout information to determine the location and nature of the obstacle, and combine it with the robot's current position, speed, and direction of movement to assess the potential impact of the obstacle on the robot's motion path; Develop obstacle avoidance and circumvention strategies based on obstacle information and the robot's motion state, including adjusting the fume exhaust path, changing the movement direction, reducing speed, or accelerating to bypass obstacles; According to the obstacle avoidance and circumvention strategies, corresponding control instructions are generated, including movement direction, speed and acceleration, and the control instructions are transmitted to the mobile part of the robot. The mobile part performs corresponding motion operations according to the instructions to realize the obstacle avoidance and circumvention functions.

8. The oil fume robot with artificial intelligence oil fume exhaust path planning according to claim 7 is characterized in that: The obstacle avoidance and circumvention strategies include adjusting the fume emission path, changing the moving direction, reducing the speed or accelerating to circumvent obstacles.

9. A computing device, characterized in that include: one or more processors; A storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, so that the one or more processors implement the operation of the robot according to any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a program, which, when executed by a processor, implements the operation of the robot according to any one of claims 1 to 8.

Images