MXPA01000730A - Robotic system - Google Patents

Abstract

A self-propelled robot is disclosed for movement over a surface to be treated. The robot has a power supply (11) and a pair of wheels (8,9) driven by motors (6, 7) for moving the robot over the ssurface. A mechanism (113, 115, 16) is provided for controllably depositing a fluent material onto the surface. Navigation sensors (4, 13, 18, 21) provide signals for enabling the robot to navigate over the surface and one or more detectors (14, 15, 17) detect the presence of the material on the surface and provide signals indicative of its presence. A control system (100) receives the signals from the sensors and detectors and controls the motors and the depositing mechanism in dependence upon the signals received from the sensors and detectors.

Description

ROBOTIC SYSTEM The present invention relates to robotic systems and, more particularly, to a mobile robotic system capable of moving on a surface and capable of treating the surface. Conventionally robotic systems, or robots, of this type can be described as semiautonomous, that is, self-propelled but that depend for navigation guidance on transmitters, receivers and sensors to establish a coordinated system by which the robot navigates, in effect learning the location of obstacles within your field of movement. More recently it has been proposed to allow a robot to move without establishing a coordinated system, rather depending on the detection of ad hoc stimuli to allow the robot to navigate around obstacles. For example, it has been proposed to provide a robotic vacuum that operates along these lines. Robotic self-navigation systems of this type are called autonomous robots. However, robots of these types, often designed for operation in a domestic environment, need a control system that is able to allow the robot to move around its environment safely and consequently, additionally need some kind of detection system of a collision that is capable of providing information about imminent collisions or collisions to a control system capable of acting very quickly to avoid collision or to minimize the impact, and to carry out the prevention of collisions by reorienting the robot before another movement . Unfortunately, the on-board processing power is inevitably limited by cost constraints in particular systems and consequently current ones, to avoid being costly in a prohibitive manner, to have relatively limiting navigation capabilities that result, in use, in the robot tracing a trajectory that involves passing over the same areas of the surface on several occasions. While this may not be problematic in, for example, a vacuum, if the robot has the function of treating the surface in other ways, then such repetitive movement may result in excessive treatment of the surface, which not only wastes the product used for treatment (a serious problem where the payload is restricted), but can also damage the surface or be otherwise harmful. The present invention is directed to provide a self-propelled robot that can overcome said problems. According to the present invention, a self-propelled robot is provided to move on a surface to be treated, the robot comprising an energy source; a traction mechanism that receives energy from the energy source, to move the robot on the surface; a mechanism for controllably depositing a fluid material on the surface; a plurality of navigation sensors that provide signals to allow the robot to navigate the surface; one or more detectors adapted to detect the presence of the material on the surface and provide signals indicating the same; and a control system that receives the signals from the sensors and detectors, to control the traction mechanism and the deposit mechanism depending on the signals received from the sensors and detectors. By detecting the application of the fluid material, which can be a liquid or gaseous fluid or a powder that can flow, excessive application of material can be avoided or minimized by either navigating the robot around already treated areas and / or controlling the deposit mechanism to stop the deposit of material on said previously treated areas. The material for preferential treatment is contained within a reservoir in the robot and may comprise compositions suitable for treatment of floors, carpets and other floor coverings. The robot can, if desired, also include means to clean the floor or floor covering before treatment, for example in the form of a vacuum device. The invention also includes a method for treating a surface using a robot as defined above. The treatment method can be used for various applications in carpets, and other floor coverings, such as cleaning, protective treatment, for example for protection against stains and dirt, fire protection, UV protection, wear resistance, control of dust mites, antimicrobial treatment and the like, as well as treatment to provide an aesthetic benefit such as odorization / deodorization. The treatment method also finds application in other surfaces such as synthetic, ceramic or wood floor coverings. As well as polishing hard surfaces, the robot can also be used to apply coatings either to improve aesthetics or to act as a protective layer. Accordingly, according to another aspect of the invention, a method is provided for controllably depositing a fluid material in floors, carpets and other floor coverings using an autonomous, self-propelled, deposit detector robot. The deposited material can, for example, be a composition for cleaning carpets, a composition for cleaning hard surfaces, or one of several compositions applied simultaneously, or consecutively, and can include a marker, whose presence can be detected for provide detection of the degree of deposit of treatment material. Said marker can have a limited detection life, for example, 12, 24 or 48 hours. The robot of the invention can also provide non-visible treatment, for example, for odor control, antibacterial action of dust mite control. The robot preferably comprises a plurality of navigation sensors that provide signals to allow the robot to navigate on the surface, one or more detectors adapted to detect the presence of the material on the surface and provide signals indicating the same. The navigation sensors may include one or more collision sensors and / or proximity sensors. The collision sensors may include one or more side shift sensors arranged in a peripheral sensor ring to provide 360 ° collision detection, and / or one or more vertical displacement sensors. Using a generally circular shape together with a control regime that explores the best escape direction after the robot has been stuck (for example, in a corner) is especially favorable. In addition, it may be additionally favorable to detect the angle of any collision, in order to optimize the rear movement angle of the robot away from the obstacle. The traction mechanism preferably includes left and right pulse wheels, coaxially arranged with corresponding pulse motors which are preferably provided with impulse signals modulated by pulse width. To deposit the material on the surface, a series of supply ports, for example, spray nozzles, may extend generally parallel with the pulse wheel axis, preferably extending to the same lateral extent of the reservoir detectors. The detectors may comprise one or more sensors arranged to detect the edge of a previously deposited product section. Suitable deposit detectors include one or more sources and / or radiation detectors, moisture detectors, reflectivity meters, conductivity meters, etc. The detectors may be arranged laterally of the driving wheels, preferably in front of them.

In addition, the robot preferably comprises a control system for controlling the deposit of the material depending on the signals received from one or more detectors and sensors. In preferred embodiments, the control system functions to control the deposit of the material (e.g., to avoid or minimize excessive application) by a combination of strategies comprising a) navigating the robot around surface areas previously treated (hereinafter referred to as the "navigation strategy", and b) controlling the deposit mechanism to stop or reduce the deposition of fluid material on the surface as the robot passes over said previously treated areas (which is hereby called the 'deposit speed control strategy'). In practice, the control system arbitrates between the two strategies depending on the signals received from the navigation sensors and deposit detectors. The ability of the control system to arbitrate between the two strategies, for example, to make a quick judgment on whether to cross or navigate around previously treated areas and whether to maintain, reduce or stop the deposit in accordance, is an important feature to ensure a controlled deposit in the context of a fully autonomous robot designed to operate in the disordered, unstructured and clutter-free environment typically found in domestic and institutional situations. Alternatively, the control system can be designed to control the tank simply by following a tank speed control strategy, in other words, controlling the tank mechanism to stop or reduce the deposit of fluid material on the surface as the robot passes over previously treated areas. Of course, systems that rely solely on deposit speed control require less complicated electronic components than the preferred combined strategy systems described above. On the other hand, single-strategy systems can be less efficient in terms of the time required to complete the task at hand. Preferably, the control system has a hierarchical architecture and includes one or more microprocessor controllers or microcontrollers to control higher level functions, and provide higher level instructions and a plurality of lower level function modules adapted to receive signals from sensors and detectors and to provide control signals in response to them. The traction mechanism and product assortment control signals are preferably output to a traction mechanism controller and a product assortment controller via a multiple or bus arranged to receive signal inputs from the microprocessor and a plurality of subprocessors each one corresponding to a respective navigation sensor or the like. By this means, a distributed processing system can be employed to provide a high level of flexibility in control strategy, while allowing a simple connection of the subprocessors, consequently to reduce the complexity and expense of the control system. The various processors of preference include neural network functionality to provide behavioral characteristics appropriate to the chosen task of the robot, the behavioral characteristics of the processors being preferably moderated by a group of generic moderators that provide the necessary arbitration between control instructions of the various processors. The functions of highest level of preference include one or more functions selected from the determination of the robot being: stuck, estimation of the size of the room, determination of the level of disorder, and battery monitoring. The lower level modules are preferably analogous neural networks that provide, for example, edge tracking and assortment control functions, together, preferably, with fall detection, collision detection, speed reduction and random movement functions . An example of a self-propelled robot constructed in accordance with the present invention, and its method of operation, will now be described with reference to the accompanying drawings in which: Figure 1 is a bottom plan view of the robot; Figure 2 is a functional diagram of the robot; and Figures 3A-C illustrate neural network aspects of part of the robot control system. As can be seen from Figure 1, the robot of the present example is substantially circular in general plan view. A simple plate-1 frame supports both the mechanical and electrical components of the robot. The plate-like frame 1 supports the body 2 of the robot on flexible rubber mounts 3 which allow the body to move relative to the frame when a force is applied, for example, by collision with an object, to a sensor ring. which is arranged around the periphery of the body. Four displacement sensors 4 placed at 90 ° intervals around the robot measure the lateral displacement of the body 2 relative to the frame 1 and inform the contact control system with an external object. The displacement sensors 4 are based on linear Hall effect devices that produce a voltage that is proportional to the strength of the magnetic field in which they were immersed. Each sensor consists of a small permanent magnet mounted on the body wrap support ring 20 and a Hall effect device mounted on the main frame 1. When the body moves with respect to the frame (as it happens during a collision) the The voltage produced by the Hall effect device varies and can be used to signal to the control system that an object has been found. By examining the signals of the four sensors, the angle and magnitude of the collision can be deduced. These sensors allow displacements in the order of 0.1 mm to be detected reliably. A fifth sensor 18, of the same type as the displacement sensors 4, measures the vertical displacement of the body shell to accommodate forces produced by objects that are of insufficient height to cause lateral body movement. In an alternative construction, these sensors can be replaced by a single sensor built according to specifications that can measure lateral and vertical displacement simultaneously. Said integrated sensor can be of an optical nature using a series of photodetectors mounted on the frame and a light source which is mounted on the body support ring. A forward-looking time-of-flight ultrasound sensor 13 is mounted on the front of the robot and is used to allow the robot to gather more information about its environment, which can be achieved by the displacement sensors 4 alone. This ultrasound sensor 13 is based on a Polaroid 6500 series sonar classification device of Polaroid® classification module, reference of Polaroid 615077, the data of which is preprocessed by a dedicated unit 5 in which the sensor 13 is located. An ultrasonic sensor unit 5, which contains the ultrasonic sensor 13 as such and a suitable electronic interface, is mounted on the body to provide proximity information to the robot control system. The left and right motors 6, 7 are provided to drive the corresponding left and right wheels 8, 9 each with a soft-smelling rim, by means of an integral reduction gearbox, to provide motive power to the robot. A single small pivoting wheel 10 mounted on the back of the robot completes the impulse / movement system and allows the frame to move forward or backward and rotate in place. Varying the rotation speed of the left and right motors 6, 7 allows the robot to be driven in any direction. The speed of the motors is controlled by modulating by pulse width the voltages applied to the motors. This involves turning the motor current on and off very fast (100,000 times per second) and varying the time ratio from 'on' to 'off' time. This is a very efficient way to control the power to the motors and consequently their speed. The energy for the robot, including the motors 6, 7 and the control system is provided by a battery pack 11 mounted on the frame 1. To protect the robot components from tampering and damage, a cover or housing (not shown) is attached to the robot. body 2 to house the robot components. In the preferred embodiment, this is a spherical part or similar to a dome. A row of spray nozzles 16 and a pump 115 (not shown in Figure 1) provide a means for dispensing the treatment fluid on the surface to be treated and the detectors 14, 15, 17 are provided to detect the presence of the treatment fluid (or a suitable additional marker fluid). The three sensor units 14, 15, 17, one placed in front of each of the pulse wheels and the third centrally positioned 17, emit light at a wavelength that excites a fluorescent dye in the product being detected. These sensor units incorporate a pair of light-sensitive devices placed at 90 ° to the direction of travel of the robot and separated by 20mm, which can detect light produced by the fluorescent dye. By examining the intensity of the light detected by these devices, the edge of a previously deposited product section can be detected and therefore monitored. In an alternative construction, the three sensor units 14, 15, 17 pass a small electrical current through the floor cover by virtue of a series of stainless steel contacts that are designed to slide over the floor covering surface. The conductivity of the floor cover will vary depending on whether it has been recently sprayed with the product or not. By examining the conductivity of the floor covering, the edge of product deposited above can be detected and consequently followed. In an alternative construction, in which fluid is going to be supplied to an edge or corner, the placement of the sprays is modified. The modification is such that the spraying is capable of delivering to the edge of the robot or beyond, for example, either by placing nozzles on the periphery of the bottom side or by additional nozzles protruding from the cover and directed in such a way as to spray beyond the perimeter of the robot. The control system of the robot comprises several circuit boards and components that are not shown in Figure 1 in detail, but which are broadly indicated by reference numerals 12 in Figure 1. The control system will now be described in more detail .

Two purposes of the control system of an autonomous mobile robot such as the one in the example are to allow the robot to move within a physical environment safely and allow it to perform useful tasks. To do this the robot must be aware of its immediate environment and be able to react to particular circumstances in particular ways. A robot designed for an unrestricted home environment needs to have certain basic skills, such as a collision detection ability, which can cause it to stop bumping into an object and then take evasive action before resuming its previous activity. In the case of collision detection, sensors 4, 18, 13, which detect impacts with and in proximity to objects, will inform the control system of the impact angle and its force. The control system must react very quickly to this stimulus and avoid any further movement in this direction. A conventional approach to this problem would be to have a computer monitor the collision sensors and act on the data to stop the motors and then perform some form of prevention maneuver. This is perfectly feasible, but if the same computer is simultaneously required to perform other tasks, for example, as in the present case, monitor other sensors and perform navigation math, there soon arrives a point where the speed and power of the computer to The required board becomes expensive in a prohibitive manner if the reaction times must be acceptable.

The alternative, adopted in the present invention, is to use discrete modules that perform functions in a manner analogous to the reflexes of a biological organism. The advantages of this system are obvious: the main processor can simply issue high level commands such as move or turn and is free to perform other abstract tasks. This alternative is a form of hierarchical distributed processing and allows the control system to be composed of simple modules that together produce faster response times than an undistributed system of the same cost. Another significant advantage of distributed processing is its inherent robustness. If a system using a conventional single processor approach suffers a failure, it can leave the system in an unsafe state, which in the case of a robot could let it crash into objects or people. The distributed approach can be designed to have a much greater degree of fault tolerance, making the occurrence of complete system failures less likely. Distributed processing can be implemented using conventional computers connected together by some form of network, but these tend to be expensive to design and implement. The approach adopted in the present invention is to stimulate biological neural networks in real analog hardware to provide a system consisting of behavioral modules, which are designed to perform individual tasks. These behaviors are handled by a simple microcontroller, which performs higher level tasks such as mathematical functions to estimate the size of the room or a strategy to escape from under a table. The control system 100 will now be described with reference to Figures 2 and 3. Figure 2 illustrates the functional relationship of the components of the control system. The control behaviors used in the robot can be divided into two basic types, low level and high level. Low-level behaviors are implemented in hardware such as discrete neural blocks or 101-105 modules, while high-level behaviors are software algorithms that run in a microcontroller 106. The functions of the low-level behavior modules 101-105 are describe now in detail: Fall - to prevent the robot from falling down the stairs is equipped with 4 fall detectors 21 that alert of vertical dangers and provide signals to the fall behavior module 101. Fall detectors 21 are infrared proximity sensors assets that comprise a modulated light source that emits a beam of infrared light directed at the target (in this case the floor), and an infrared detector that monitors the intensity of the light that is reflected. When the sensor is directed over a falling edge, the intensity of the reflected light decreases and the sensor informs the hazard control system. This behavioral function has a very high priority and when active it operates to maneuver the robot away from the danger and return it to a course that is modified to avoid falls. Edge tracking - edge tracking module 104 provides a behavior function that uses information from sensors 14, 15, 17 that allow the robot to find the edge of a previously treated area (as described above) and move to along that edge to produce a faster exploration of the floor surface. Random - in the absence of any edges the robot moves in a random direction under the action of a random motion module 114 until an object is found or until the edge tracking behavior is activated. Collision - the collision detection module 102 enters from the displacement sensors 4, 18 and operates in such a way that when encountering an obstacle the robot stops, it moves in reverse a small distance, then turns away from the object in a direction depending on the angle of impact, which is determined from the signals of the displacement sensors 4, 18. Speed reduction - when an object is detected by the ultrasound sensor unit 5 within a predetermined range limit, the forward speed of the robot is reduced by the Reduction module of Speed 103 to minimize the impact force generated when contact with the object occurs.

Assortment - an assortment control module 105 has inputs from a fluid level sensor 203 and sensors 14, 15, 17 by the Edge Tracking module 104. If the UV sensors 14, 15, 17 report an untreated carpet in the direction of travel the treatment chemical is stocked until the treated areas are found or the fluid level reaches a lower limit. The high level behaviors are determined within the microcontroller 106 and comprise the following functional modules: Jammed - a routine 107 determines whether there has been more than a chosen number of collisions in a selected period and causes the robot to stop and use the search engine. Ultrasonic range 5, 13 to find the longest clear route and move in that direction. The robot will rotate in place, operating wheels 8, 9 in opposite directions, looking for the longest clear route. When the best address is discovered the robot will move in that direction. Estimation of the size of the room - using the collected statistics of the ultrasound sensor 13 and measuring the time between collisions, the routine 108 is able to estimate the area of the room. This is used to determine how long it would take the robot to treat a particular room. Estimation of the level of disorder - comparing estimates of the size of the room against collisions per minute a routine 109 is able to deduce a factor that describes the complexity of the room. This can then be used to modify the operating time to consider the level of clutter. Battery monitor - a battery monitor routine 110 checks the battery status by monitoring the output and current voltage. Use this information to estimate how long the battery will be able to withstand the robot's systems before a recharge is needed. When the monitor routine decides that the battery status is approaching the point where reliable operation is no longer possible, the user is warned by lighting a low battery indicator. If the robot is allowed to continue operating without being recharged the monitor routine will shut down the robot in a safe and controlled manner when the energy levels reach a predetermined point. Nickel-cadmium or nickel metal hydride batteries require careful charging to ensure maximum capacity and life time and the monitor routine also controls the battery charge cycle to ensure these needs are met. Traditionally, neural network designers have insisted that each neuron in a network be connected to another neuron in that network. While this allows the network the highest level of flexibility, many (as much as 90%) of these connections will never be used. The system of the present invention allows preconfigured neural networks to be connected together in a much less complex manner allowing the behavior of the robot to dynamically adjust to the immediate environment in a continuous manner.

This so-called "multiple architecture" comprises an analogous bus or multiple 111, which connects all the behavioral modules 101-105 and their associated actuators with each other. Four generic moderators arbitrate between the behaviors, and give rise to a prototype behavior of them that regulates the general activity of the robot by means of a motor controller 112 and assortment fluid pump controller 113 that drives the pump 115. These generic moderators add up all the excitatory and inhibitory inputs and apply a non-linear transfer function to the results. The outputs of these moderators form the inputs to the motor controllers. In order to explain the function of the multiple architecture, it is necessary to describe the basic neural aspects of the control system. For this purpose, reference will be made to Figures 3A-C. A single neuron (see figure 3A) has three basic types of connections, excitatory inputs that cause the neuron to 'turn on', inhibitory inputs that suppress activity and output that represents the state of the neuron. In addition, neurons may have other properties such as mposition which causes the output to fall slowly over time, and Threshold that suppresses all output until the sum of the entire input exceeds a certain level. Figure 3B shows (by example) a simplified representation of the collision behavior and the multiple system in neural notation.

The collision sensors 4 are represented in FIG. 3B as 1, 2, 3 and 4 and are damped and normalized by sensor preprocessors 5, 6, 7 and 8. The outputs of the sensor preprocessors are each fed into a single neuron 9, 10, 11 and 12 configured as a pulse stretcher with a time constant of approximately 5 seconds. The outputs of these neurons are connected to the rest of the network formed by neurons 13 to 28 where the pattern of connections, and transfer characteristics of the neurons give rise to the behavior as such. The outputs of this network are connected through connections 41 to 48 to the multiple adders (generic moderators) 29 to 32 where the signals are summed and outputs 37 to 40 form the inputs to the left and right motor controllers (not show in this figure). Connections of other unspecified behavior (of which there may be many) are shown as 50 to 57. Connection 49 is a subsuming entry, which is used to disable the entire behavior under control of the programmer software running on a microcontroller or another neural behavior of higher priority. The sensor outputs are also available to the microcontroller so that high-level behaviors such as clutter level estimation can have access to whatever data is produced. In the case of a direct collision while moving forward, the following is true: the previous collision sensor 1 produces a pulse as a contact when an obstacle occurs. This pulse is amplified by the sensor preprocessor element 5 and passed to the input neuron 9. This neuron is configured to stretch the width of an input pulse (when that pulse exceeds a predetermined input threshold) to approximately 5 seconds. The output of the input neuron 9 is fed simultaneously to four other neurons 13, 14, 15 and 16. These 'hidden layer' neurons are configured to act as attenuators or in 'weighting' neural terms, and therefore change the amplitude of the signals applied. The neurons 13 and 15 are set to produce an output level of 10 (maximum) when they are excited and the outputs are connected to the output neurons 22 and 26 which, when excited, apply signals to the manifold giving instructions to the motors to stop the movement forward. The neurons 14 and 16 are set to produce an output of 5 (half) when they are excited and their outputs are connected to the output neurons 23 and 27 which, when excited, apply signals to the manifold giving instructions to the motors to move the robot backwards. This part of the behavior as such, would theoretically lead to a situation where the robot would repeatedly collide and withdraw in a straight line from the obstacle, but inaccuracies inherent in the control system and impulse mechanics together with the fact that the probability of a perfect direction during The collision is remote, meaning that the other collision strategies involving the left and right sensors, will cause the robot to turn as it moves in reverse of an obstacle and produces a useful behavior.

The manifold function will now be described in detail with reference to Figure 3C. The manifold as its name implies gathers all the output from the various neural behaviors of the robot, the sum and provides the inputs to the motor controllers. Figure 3C shows the section that controls the right motor controller; and the left section is identical. Connection 41 is effectively the 'Go forward' entry and 42 is 'Do not go forward'. These two opposite inputs are fed into the excitatory and inhibitory inputs of the neuron 29. If the values of Go forward 6 and Do not go forward 3 are applied simultaneously, the neuron 29 produces a value of 3, but if the values they are inverted, ie Go forward 3 and Do not go forward 6, neuron 29 produces 0. This is very important since it allows a behavior to inhibit movement in a particular direction without causing movement in the opposite direction. Neuron 30 performs the same task as 29 except that its inputs are 'Go backwards' 43 and' Do not go backwards' 44. Neuron 29 is connected to the exciter input of 33 which in turn drives the 'Go to ahead of the right motor controller via connection 37. Neurons 30 and 34 are connected to the 'Go backward' input of the right motor controller via connection 38. The motor controller sum these inputs so that Go forward 8 and Go backwards 4 applied simultaneously at connections 37 and 38 respectively result in the right wheel rotating forward at a speed of 4. Neurons 33 and 34 also have inhibitory connections where the forward signal path is connected to the reverse path and vice versa. This allows a non-linear behavior of the multiple and as the resistance of these connections increases, the robot is less likely to enter a stable state, where no movement occurs due to behaviors with conflicting interests that are imposed simultaneously. Now more details of some of the various sensors and their operation will be given: The ultrasound sensor unit 5 has a preprocessor that handles the sensor 13, providing pulses timers, etc., and provides the high level behavior with data of 'reach to the objective 'continuous' and a simple-range warning to the speed reduction behavior module 103. The continuous output is used by the jamming behavior module 107 which rotates the robot 360 ° while looking for a clear path by where the robot can escape and is also used by the behavioral modules for estimating the room size and disorder estimate 109, 108. To perform the task of dispensing the treatment compositions (e.g., a carpet cleaning formulation, known per se, comprising an aqueous solution of anionic surfactant, optionally together with a polycarboxylate dirt suspending agent) in an surface, it is desirable to know what areas of the surface have already been treated. A labeling agent, added to the formulation in question, has characteristic properties such as absorption or emission of light at a known frequency, or fluorescent behavior that can be detected by the robot. Examples of such labels are luminol, which can be reacted with hydrogen peroxide to emit light, and substituted coumarins such as 7-hydroxy or 4-methyl-7-hydroxy variants which are highly fluorescent but undergo ring opening reactions for form a non-fluorescent derivative. For detection purposes, a corresponding light source and photodiode detectors 14, 15, 17 are placed on the left and right in front of the pulse wheels 6, 7 of the robot to detect said marker chemical and allow the system control follow the edge of a previous pass. In this way, a structured assortment pattern can be established. In addition, the detector can be linked, by means of a negative feedback system, to the assortment arrangement, thus avoiding the deposition of formulation in an area of the surface that has already been treated. When a floor area that has not been treated can not be found, the actual time taken is compared to data provided by the estimated room size behavior module 108, and if the two are within acceptable limits, the treatment of the floor is considered complete. The characteristic properties by which the marker is detected are decomposed in the 24-48 hours following the application (by oxidation area or photolytic decomposition) or, in the case of a two-stage treatment method, a second chemical can be applied to the first, neutralizing the characteristic properties of the chemical marker. An alternative means to achieve this desired behavior is to use moisture detection to identify areas of the surface that have already been treated. In this case, the inherent moisture of a liquid formulation is used to detect surfaces treated by moisture sensing arrangements that are placed to the left and right in front of the drive wheels 6, 7 of the robot. Again, this system can be used to allow the robot to follow the edge of a previous pass. In cases where a hard floor surface is being treated (for example, with an aqueous cleaning formulation comprising a medium nonionic chain length surfactant with citrate-carbonate and caustic soda) the reflective properties of the floor can be used to detect what areas of the floor have been treated. A high-intensity light source directs light onto the floor where, following the reflection, it is subsequently detected by a photodiode detector. These are placed on the left and right in front of the drive wheels 6, 7 of the robot. Again, this system can be used to allow the robot to follow the edge of a previous pass. In this case, the ability of a formulation to reduce the reflection capacity of the floor is exploited to allow its detection.

Claims (11)

NOVELTY OF THE INVENTION CLAIMS

1. - A self-propelled robot to move on a surface to be treated, the robot comprising an energy source (11); a traction mechanism (6-9) that receives energy from the power supply, to move the robot on the surface; a mechanism (16) for controllably depositing a fluid material on the surface; a plurality of navigation sensors (4, 13, 18, 21) that provide signals to allow the robot to navigate the surface; one or more adapted detectors (14, 15, 17) to detect the presence of the material on the surface and provide signals indicating the same; and a control system (12, 100) that receives the signals from the sensors and detectors, to control the traction mechanism and the reservoir mechanism depending on the signals received from the sensors and detectors.

2. The robot according to claim 1, further characterized in that the navigation sensors include collision sensors comprising one or more lateral displacement sensors (4) arranged in a peripheral sensor ring (20) to provide collision detection 360 °, and / or one or more vertical displacement sensors (18).

3. The robot according to claim 1 or claim 2, further characterized in that the detectors comprise one or more sensors (14, 15) arranged to detect the edge of a section of product deposited previously.

4. The robot according to any of claims 1 to 3, further characterized in that the deposition detectors include one or more sources and / or radiation detectors, humidity detectors, reflectivity meters, conductivity meters.

5. The robot according to any of claims 1 to 4, further characterized in that the control system (100) has a hierarchical architecture and includes one or more microprocessor controllers or microcontrollers (106) to control higher level functions. and provide higher level instructions; and a plurality of lower level function modules (101-104, 114) adapted to receive signals from the sensors and detectors (4, 13-15, 17, 18, 21) and having processors to provide control signals in answer to them.

6. The robot according to claim 5, further characterized in that the signals of the traction mechanism and control signals of product assortment are emitted to a traction mechanism controller (112) and to a product assortment controller (113). ) by means of a multiple or bus (111) arranged to receive signal inputs from the microprocessor (s) or microcontroller (s) (106) and from the lower level function modules (101-104).

7. - The robot according to claim 5 or claim 6, further characterized in that the lower level function module processors (101-104, 114) include neural network functionality to provide appropriate behavioral characteristics for the chosen task of the robot , the behavioral characteristics of the processors being moderated by a group of generic moderators (111, 29-32) that provide arbitration between the control instructions of the various processors (101-104, 114).

8. The robot according to claim 7, further characterized in that the lower level modules (101-104, 114) comprise analogous neural networks that provide functions selected from the functions of edge tracking and assortment control, functions of Fall detection, collision detection, speed reduction and random movement.

9. A method for controllably depositing a fluid material on floors, carpets and other floor coverings that uses an autonomous, self-propelled robot, deposit detector and reservoir speed controller.

10. The method according to claim 9, further characterized in that the deposited material is a carpet cleaning composition., an odorization / deodorization composition, a dust mite control composition, an antimicrobial composition, a hard surface cleaning composition, or one of several compositions applied simultaneously, or consecutively.

11. The method according to claim 9 or claim 10, further characterized in that the deposited material includes a marker, whose presence can be detected to provide detection of the degree of deposit of treatment material.