A multi-sensor cleaning system that optimizes cleaning parameters based on interdependent principles according to surface conditions.

Abstract

The present invention relates to robotic cleaning systems, combined analysis of multiple sensor data, control systems based on sensor fusion, and optimization of cleaning parameters, and specifically concerns a cleaning system that determines the surface condition using multiple sensor data and, based on this determination, enables the interdependent optimization of cleaning parameters.

Description

1 TARIFF 5 CLEANING PARAMETERS ACCORDING TO SURFACE CONDITION MULTIPLE SENSORS THAT OPTIMIZE THROUGH INTERDEPENDENCE CLEANING SYSTEM Technological Field: This invention involves robotic cleaning systems, combined analysis of multiple sensor data, sensor 10 fusion-based control systems and optimization of cleaning parameters and, in particular, the determination of surface condition using multiple sensor data and this determination depending on the interdependent optimization of cleaning parameters It is related to a cleaning system that provides State of the Art: 15 Robotic cleaning systems and mop systems are widely used today. Cleaning devices generally have limited detection mechanisms based on surface type. It is in operation and the cleaning parameters are fixed or predefined modes. It adjusts within this framework. The sensor structures used in these systems are mostly singular. It is aimed at measuring parameters, for example, only the surface type or obstacle 20 It focuses on determining the condition and the immediate physical state of the surface and the dirt. It is unable to comprehensively evaluate its characteristics. Some of the existing solutions... Advanced systems offer limited adaptive control based on specific sensor data. While these systems can offer mechanisms, sensor data in these systems are mostly independent. It is considered as such and a combined decision model representing the surface situation 25 This is not the case. This situation means that data from different sensors are combined to form a single image. Due to the inability to interpret the density, moisture content, or stickiness of the dirt on the surface. This leads to the inability to correctly classify critical parameters such as characteristics. It opens up. In addition, cleaning parameters in existing systems, such as vacuum power, water The delivery rate, brush pressure, and device movement speed are generally independent of each other or 30 These parameters are adjusted within a limited interaction and are interdependent. This is not taken into consideration. Therefore, the systems are optimal for different surface and dirt conditions. inability to adapt in this way, sometimes resulting in excessive water or energy consumption. While this is happening, in some cases inadequate cleaning performance occurs. In addition, in existing technical solutions, sensor 35 is obtained during the cleaning process. data is continuously evaluated within a feedback mechanism for cleaning. An efficient closed-loop control for dynamically updating its parameters. 2 There is no such approach. Existing systems are mostly open-loop or limited feedback. It operates with fed-control structures, monitoring the surface condition before and after cleaning. No performance evaluation is being conducted. Therefore, the current technical... The solutions are a surface analysis based on the combined evaluation of multiple sensor data. its approach involves optimizing cleaning parameters in an interdependent manner, and Holistic control that provides continuous adaptation together with closed-loop feedback 10 The system is unable to deliver. These shortcomings affect cleaning effectiveness, energy and water efficiency. This presents significant limitations in terms of adapting to different surface conditions. The purpose of the invention: The purpose of this invention is to determine the physical condition and dirt characteristics of the surface to be cleaned. By determining vacuum power, water flow rate, and brush pressure through combined analysis of multiple sensor data. equalizing cleaning parameters such as parameters and movement speed in a dependent manner. The goal is to provide a cleaning system that optimizes the process over time. This allows for cleaning different surfaces. and adapts to dirt and grime conditions in real time, increasing cleaning efficiency. a system that reduces energy and water consumption and minimizes the need for user intervention A control mechanism is obtained. 20 Explanation of the Figures Figure 1 - Schematic showing the general structure and sensor placement of the smart cleaning system. appearance Figure 2 - Flow diagram illustrating the adaptive cleaning control process based on surface analysis. view 25 References: 1. Cleaning device body 2. Vacuum motor 30 3. Water tank 4. Water pump 5. Brush mechanism 6. Humidity sensor 7. Optical dirt detection sensor 35 8. Surface type detection sensor 9. Pressure / resistance sensor 3 10. Microcontroller unit 5 11. Data processing module 12. Surface analysis algorithm 13. Cleaning control algorithm 14. Adaptive control system 15. User interface 10 16. Wireless communication module 17. Power supply 18. Timer unit 19. Regional mapping module 20. Safety and fault control system 15 Description of the Invention: The invention uses multiple sensors to determine the physical condition and dirt characteristics of the surface to be cleaned. Determining through combined analysis of data and cleaning based on this determination. a cleaning process that simultaneously optimizes its parameters in a dependent manner. It relates to the system. Unlike existing systems, in this invention, sensor data is 20 It is not evaluated independently, but rather data obtained from different sensors. combined within a weighted decision model to represent the surface state. A unique surface condition score (S) is obtained. Additionally, cleaning parameters are also included. not independently, but interdependent depending on the surface condition score This is determined simultaneously within the scope of the optimization model. This approach, 25 by taking into account the interactions between parameters, it provides a solution to classical systems. achieving higher cleaning efficiency and lower energy and water consumption compared to others. This structure makes it possible to evaluate sensor data individually. Unlike systems that rely on interaction between cleaning parameters By enabling this to be determined, a technical improvement in system performance is achieved. 30 It constitutes. The system is located inside the cleaning device housing (1) and the vacuum motor (2), water pump (4) working in connection with water tank (3), brush mechanism (5) and this sensor, control and energy management subsystems that interact with components It consists of: Surface condition, moisture sensor (6), optical dirt detection sensor (7), surface type 35 Simultaneous data generated by the sensing sensor (8) and the pressure / resistance sensor (9) This is determined by collecting the sensor data. The obtained sensor data is then used by the microcontroller. 4 The data is processed within the data processing module (11) managed by unit (10), and this 5 Within this scope, sensor data undergoes noise filtering and min-max It is scaled to the [0–1] range using the normalization method. Normalized sensor data is transferred to the surface analysis algorithm (12) and by this algorithm the following Surface state score (S) using weighted decision function It is calculated as: 10 S = σ(w₁N + w₂K + w₃R + w₄T) Here, N represents the moisture content, K the dirt density, R the surface resistance, T the surface type, and w₁–w₄. It represents the weighting coefficients. The σ function is a nonlinear transformation. It is a function and has a sigmoid or piecewise linear structure. The resulting surface condition score... (S) is classified according to predefined threshold values, determining the surface dirt characteristic. 15 This is determined. In this context, if S < 0.3, it is considered low-density dry dirt. If 0.3 ≤ S < 0.7, it indicates moderate dirt, and if S ≥ 0.7, it indicates adhesiveness. The level of dirt is being assessed. Surface condition obtained by surface analysis algorithm (12), cleaning control The algorithm (13) is transferred to this algorithm and the cleaning parameters are taken as one parameter 20 It is considered within the scope of the vector (P). The parameter vector in question is vacuum power (V), The parameters include water flow rate (W), brush pressure and rotational speed (B), and device movement speed (H). The cleaning control algorithm (13) does not treat these parameters independently, It is treated as a mutually dependent optimization problem and the following objective It determines simultaneously according to its function: 25 F = αE + βC + γR_c Here, E represents energy consumption, C represents water consumption, and R_c represents the amount of dirt remaining after cleaning. It represents the system priorities, and the coefficients α, β, and γ are determined according to the system priorities. The parameter vector (P) refers to the surface condition score and the feedback obtained after cleaning. It is updated based on the data using the iterative error minimization principle. This 30 The difference between surface conditions before and after cleaning in this context: e = S_before − S_after It is defined in this way, and parameters are rule-based using this error value. It is optimized through update or gradient-like enhancement steps. For example, the simultaneous detection of high surface resistance and moderate moisture content is 35 In this case, the system takes into account this situation, which indicates the presence of sticky dirt, and vacuums. by increasing its power, increasing the water flow in a controlled manner, and increasing the brush pressure. It optimizes cleaning efficiency. In contrast, low humidity and low dirt density 5 In this case, vacuum power and water usage are used to save energy. is reduced. The system is closed-loop control within the scope of adaptive control system (14). It operates according to the principle that sensor data is continuously monitored throughout the cleaning process. This allows the system to react to changes in surface conditions in real time. It can dynamically update cleaning parameters by adapting to the environment. 10 The regional mapping module (19) divides the cleaning area into sub-regions, and each region It creates a sensor data history for the timer unit (18). Initial cleaning parameters for different regions can be adapted by evaluating them. This enables the determination of weighting coefficients (w₁–w₄) and optimization parameters. Based on past cleaning performance, the principle of minimizing errors is applied. 15 This process is being updated so that the system adapts to different surface and dirt conditions over time. It enables it to acquire an adaptable structure. The wireless communication module (16) enables the system to communicate with external data sources. by enabling parameter updates and software improvements remotely. It enables the implementation. Through the user interface (15), the user 20 Preferences are transmitted to the system, and these preferences are weighted in the optimization process. It contributes to the determination of the coefficients. The power supply (17) is used to determine the system components. by providing energy and measuring instantaneous energy consumption, it supports the optimization process. It provides data on the energy parameter used. Safety and fault control system (20) by detecting abnormal situations that may occur in the system components, the system is kept safe. 25 This structure allows the system to combine data from multiple sensors into a working mode. By evaluating it, it determines the surface condition with high accuracy, and cleansing. optimizing its parameters in an interdependent manner and closed-loop feedback Its mechanism continuously improves cleaning performance. Industrial Application of the Invention 30 This invention is compatible with robotic cleaning systems and industrial cleaning machines. It is designed to be integrated with existing production infrastructures. It is of a quality. The humidity sensor (6) and optical dirt detection sensor used within the system. Components such as (7), surface type sensing sensor (8) and pressure / resistance sensor (9) are available on the market. This can be achieved with commonly available sensor technologies and 35 easily on the microcontroller unit (10) and data processing module (11) It is applicable. The surface analysis algorithm (12) and cleaning described in the invention. 6 control algorithm (13) is structured to run on embedded systems 5 and can be integrated into existing robot vacuum cleaner platforms via software updates. This allows the system to be used not only in newly manufactured devices, but also in existing ones. It can also be used to improve the performance of devices. The system uses vacuum. with standard cleaning components such as motor (2), water pump (4) and brush mechanism (5) They work seamlessly and the need for additional hardware is kept to a minimum. This 10 The situation is that cleaning performance can be improved without creating a significant increase in production costs. This makes it possible to increase the wireless communication module (16) through the system. It can be continuously improved through remote software updates and parameter enhancements. It has a structure. The system is provided with parameter settings via the user interface (15). It can be adapted to different usage scenarios. The invention is not limited to just household robots. Not limited to vacuum cleaners, but also used in shopping malls, hospitals, airports and industrial settings. It can also be applied to autonomous or semi-autonomous cleaning systems used in facilities. It is of a quality. Thanks to the regional mapping module (19), different cleaning can be done in large areas. The implementation of these strategies becomes possible, and operational efficiency is increased. With these features, the invention offers low integration costs, compatibility with existing systems, and 20 Thanks to its wide range of applications, it offers a directly applicable solution in industry.

Claims

7 REQUIREMENTS 5 1. The present invention relates to robotic cleaning systems and the combined analysis of multiple sensor data. control systems and cleaning parameters based on sensor fusion It relates to optimization, particularly of surface condition using multiple sensor data. determination and the interaction of cleaning parameters based on this determination. It relates to a cleaning system that enables dependent optimization, 10 Features; Cleaning device body (1), Vacuum motor (2), Water tank (3), Water pump (4), Brush mechanism (5), Moisture sensor (6), Optical dirt detection sensor (7), Surface type sensing sensor (8), Pressure / resistance sensor (9), Microcontroller unit (10), Data processing module (11), Surface analysis algorithm (12), Cleaning control algorithm (13), Adaptive control system (14), User interface (15), 15 Wireless communication module (16), Power supply (17), Timer unit (18), With the regional mapping module (19) and the safety and fault control system (20) It is characterized.

Images