Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
METHOD OF CORRECTION OF ODOMETRY ERRORS DURING THE AUTONOMOUS DRIVE OF A WHEEL-EQUIPPED APPARATUS AND CONTROL UNIT THEREOF
Document Type and Number:
WIPO Patent Application WO/2023/002405
Kind Code:
A1
Abstract:
The present invention relates to a method of correction of odometry errors during the autonomous drive of a wheel-equipped apparatus (100) comprising drive means (120) operatively connected to at least two drive wheels (121, 122), sensor means (125,126) operatively connected to said at least two drive wheels (121, 122) and a control unit (200) operatively connected to said drive means (120) and to said sensor means (125,126), said method comprising: - an acquisition phase, wherein said sensor means (125,126) acquire at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122); - a generation phase, wherein said control unit (200) generates a corrective signal that controls said drive means (120) in such a way as to independently operate said at least two drive wheels (121, 122) on the basis of said at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122).

Inventors:
BERTAIA ANDREA (IT)
Application Number:
PCT/IB2022/056705
Publication Date:
January 26, 2023
Filing Date:
July 20, 2022
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
ALBA ROBOT S R L (IT)
International Classes:
G01C22/02; A61G5/04; B60K6/445; B60K23/04; B60K28/16; B60T8/175; B62D11/02
Foreign References:
US5387031A1995-02-07
US5117934A1992-06-02
US6447076B12002-09-10
US5109339A1992-04-28
Attorney, Agent or Firm:
CAMOLESE, Marco et al. (IT)
Download PDF:
Claims:
CLAIMS

1. Method of correction of odometry errors during the autonomous drive of a wheel- equipped apparatus (100) comprising drive means (120) operatively connected to at least two drive wheels (121, 122), sensor means (125,126) operatively connected to said at least two drive wheels (121, 122) and a control unit (200) operatively connected to said drive means (120) and to said sensor means (125,126), said method comprising:

- an acquisition phase, wherein said sensor means (125,126) acquire at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122);

- a generation phase, wherein said control unit (200) generates a corrective signal that controls said drive means (120) in such a way as to independently operate said at least two drive wheels (121, 122) on the basis of said at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122).

2. Method according to claim 1, wherein, during said generation phase, said corrective signal controls said drive means (120) in such a way as to generate an angular velocity difference between said at least two drive wheels (121, 122) on the basis of said at least one signal representative of mechanical vibrations of at least one drive wheel, and wherein said angular velocity difference minimizes a differential value between a current position and a predefined position of said wheel-equipped apparatus (100).

3. Method according to claim 2, wherein said control unit (200) determines the angular velocity difference between said at least two drive wheels (121, 122) in accordance with a predefined function having as arguments said at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122).

4. Method according to claim 2 or 3, wherein said control unit (200) determines the angular velocity difference between said at least two drive wheels (121, 122) in accordance with an output signal of a previously trained neural network having as input signal said at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122).

5. Control unit (200) for correcting odometry errors during the autonomous drive of a wheel- equipped apparatus (100) comprising drive means (120) operatively connected to at least two drive wheels (121, 122) and sensor means (125,126) operatively connected to said at least two drive wheels (121, 122) and adapted to acquire at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122), said control unit (200) being operatively connected to said drive means (120) and to said sensor means (125,126), and being adapted to generate a corrective signal for controlling said drive means (120) in such a way as to independently operate said at least two drive wheels (121, 122) on the basis of said at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122).

6. Control unit (200) according to claim 5, wherein said corrective signal is adapted to control said drive means (120) in such a way as to generate an angular velocity difference between said at least two drive wheels (121, 122) on the basis of said at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122), said angular velocity difference being adapted to minimize a differential value between a current position and a predefined position of said wheel-equipped apparatus (100).

7. Control unit (200) according to claim 6, wherein said control unit (200) is adapted to determine the angular velocity difference between said at least two drive wheels (121, 122) in accordance with a predefined function having as arguments said at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122).

8. Control unit (200) according to claim 6 or 7, wherein said control unit (200) is adapted to determine the angular velocity difference between said at least two drive wheels (121, 122) in accordance with an output signal of a previously trained neural network having as input signal said at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122).

9. Autonomous-drive wheel-equipped apparatus (100) comprising drive means (120) operatively connected to at least two drive wheels (121, 122), at least one pivoting wheel (123,124) operatively connected to sensor means (125,126) and a control unit (200) operatively connected to said drive means (120) and to said sensor means (125,126), said wheel-equipped apparatus (100), wherein said sensor means (125,126) are adapted to acquire at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122), and wherein said control unit (200) is adapted to generate a corrective signal for controlling said drive means (120) in such a way as to independently operate said at least two drive wheels (121, 122) on the basis of said at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122).

10. Wheel-equipped apparatus (100) according to claim 9, wherein said corrective signal is adapted to control said drive means (120) in such a way as to generate an angular velocity difference between said at least two drive wheels (121, 122) on the basis of said at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122), said angular velocity difference being adapted to minimize a differential value between a current position and a predefined position of said wheel-equipped apparatus (100).

11. Wheel-equipped apparatus (100) according to claim 10, wherein said control unit (200) is adapted to determine the angular velocity difference between said at least two drive wheels (121, 122) in accordance with a predefined function having as arguments said at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122).

12. Wheel-equipped apparatus (100) according to claim 10 or 11, wherein said control unit (200) is adapted to determine the angular velocity difference between said at least two drive wheels (121, 122) in accordance with an output signal of a previously trained neural network having as input signal said at least one signal representative of mechanical vibrations of at least one drive wheel (121, 122).

Description:
METHOD OF CORRECTION OF ODOMETRY ERRORS DURING THE AUTONOMOUS DRIVE OF A WHEEE-EQUIPPED APPARATUS AND CONTROE UNIT THEREOF

DESCRIPTION

The present invention relates to a method of correction of odometry errors during the autonomous drive of a wheel-equipped apparatus, in accordance with the preamble of claim 1. In particular, a method of correction of odometry errors during the autonomous drive of a wheel-equipped apparatus and relative control unit is illustrated.

The field of use of the present invention concerns the autonomous drive of wheelchairs, carriages, stretchers, trolleys or similar wheel-equipped apparatuses in enclosed environments such as for example hospitals, airports, shopping centres and so on, for which current geolocation systems are ineffective.

Currently, autonomous-drive systems for wheel-equipped apparatuses are known, such as for example the system indicated in US patent application US2017266069, in which an autonomous-drive wheelchair is described.

These systems can be used for the movement of people and/or things within enclosed environments in which geolocation systems, such as for example the GPS and/or GALILEO system, are not effective for locating the wheel-equipped apparatuses in the aforesaid environments.

This is due to the strong attenuation suffered by the electromagnetic signals used by such geolocation systems within enclosed environments.

To overcome this inefficiency of the geolocation systems within enclosed environments, autonomous-drive systems for enclosed environments can use odometry techniques. These techniques make it possible to estimate the position of an autonomous-drive wheel-equipped apparatus based on information coming from sensors that measure, during a time of movement of the wheel-equipped apparatus, the space travelled by some of the wheels of the wheel-equipped apparatus itself starting from a known initial position.

Such odometry techniques turn out to be very sensitive to the errors introduced in the measurements of the sensors during the period of movement of the wheel-equipped apparatus itself, for example such errors may be relevant at the beginning of the measurements of the sensors, when the wheel-equipped apparatus is in the initial position, in particular in the case where the wheel-equipped apparatus is moved on a smooth, or slippery, surface, for which one or more wheels of the wheel-equipped apparatus may be subject to slippage.

The object of the present invention is therefore to solve these and other problems of the prior art, in particular to indicate a method, and a relative control unit, of correction of odometry errors during the autonomous drive of a wheel-equipped apparatus in enclosed environments, without the aid of geolocation systems.

Another object of the present invention is to indicate a method, and a relative control unit, of correction of odometry errors during the autonomous drive of a wheel-equipped apparatus on a smooth or slippery surface.

A further object of the present invention is to indicate a method, and a relative control unit, of correction of odometry errors during the autonomous drive of a wheel-equipped apparatus that allow to minimize a differential value between a current position and a predefined position of the wheel-equipped apparatus.

The invention described consists of a control unit of an autonomous-drive wheel-equipped apparatus, and the relative method, of correction of odometry errors during the autonomous drive of the apparatus itself in such a way as to consider, in real time, the slippage of one or more wheels of the wheel-equipped apparatus itself.

Further advantageous features of the present invention are the subject of the accompanying claims which form an integral part of the present disclosure.

The invention will be described in detail below through non-limiting examples of embodiments with particular reference to the accompanying figures, in which:

- Figure la schematically represents an example of an autonomous-drive wheel-equipped apparatus according to an embodiment of the present invention;

- Figures lb and lc schematically represent an example of a drive wheel of the autonomous-drive wheel-equipped apparatus of Figure la;

- Figure 2 represents an exemplary block diagram of a control unit of the autonomous- drive wheel-equipped apparatus in Figure la;

- Figure 3 represents an example flowchart of a method of correction of odometry errors during the autonomous drive of the wheel-equipped apparatus in Figure la.

With reference to Figure la, an autonomous-drive wheel-equipped apparatus 100 is schematically represented. Such a wheel-equipped apparatus 100 may comprise drive means 120 operatively connected to at least two drive wheels 121, 122, sensor means 125, 126, operatively connected to the drive wheels 121, 122 and a control unit 200 operatively connected to the drive means 120 and to the sensor means 125, 126.

For example, in the present embodiment of the invention, the autonomous-drive wheel- equipped apparatus 100 may comprise a first drive wheel 121 and a second drive wheel 122, operatively connected to the drive means 120 by mechanical elements 127, such as for example gears, belts, drive axles, bearings, and so on.

The drive means 120 may comprise for example a first electric motor operatively connected to the first drive wheel 121 and a second electric motor operatively connected to the second drive wheel 122.

The control unit 200 is adapted to control the drive means 120 to carry out the autonomous drive of the wheel-equipped apparatus 100. The control unit 200 may generate a control signal of the drive means 120, which may be represented by an analogue or digital signal. This control signal can be transmitted to the drive means 120 via a data/power bus 101 adapted to operatively connect the control unit 200 to the drive means 120. For example, the control signal may operate the first electric motor and/or the second electric motor, in such a way that each drive wheel may be driven independently of the other. Further, the wheel- equipped apparatus 100 may comprise at least one pivoting wheel 123.

The sensor means 125, 126, are adapted to acquire at least one signal representative of mechanical vibrations of at least one drive wheel 121, 122. The sensor means 125, 126 may for example comprise at least one of the following sensors: a piezoelectric sensor, an acoustic sensor, a capacitive sensor, an inductive sensor, and a resistive sensor.

The signal representative of mechanical vibrations of at least one drive wheel 121, 122 may be of analogue or digital type and may be transmitted as an output from the sensor means 125, 126 to the control unit 200 via the data/power bus 101, adapted to operatively connect the control unit 200 to the sensor means 125, 126.

For example, in the present embodiment of the invention, first sensor means 125 are operatively connected to the first drive wheel 121, while second sensor means 126 are operatively connected to the second drive wheel 122.

With reference to Figure lb, the first drive wheel 121 is represented, resting on a rest surface 190 of the wheel-equipped apparatus 100, operatively connected to the drive means 120 by means of the mechanical elements 127, wherein the first sensor means 125 can be operatively connected to the mechanical elements 127. The same configuration (not shown in Figure lb) may be adopted for the second drive wheel 122 with the second sensor means 126. Additionally or alternatively, with reference to Figure lc, the first drive wheel 121 is represented, resting on the rest surface 190, operatively connected to the drive means 120 by means of the mechanical elements 127. The first sensor means 125 may be located at a predefined distance from a contact surface 191 between the first drive wheel 121 and the rest plane 190. This predefined distance can be, for example, comprised between 1mm and 1cm. The same configuration (not shown in Figure lc) may be adopted for the second drive wheel 122 with the second sensor means 126.

This advantageously makes it possible to detect the mechanical vibrations due to the rolling of at least one drive wheel 121,122 in contact with the excessively smooth or slippery rest surface 190 on which the wheel-equipped apparatus 100 moves.

In an embodiment of the invention, the wheel-equipped apparatus 100 may optionally comprise command means 110 (Figure la) adapted to control the drive means 120, wherein the command means 110 are adapted to be manually operated by a user. The command means 110 may comprise, for example, joysticks, buttons, and so on. The command means 110 may be operatively connected to the drive means 120, for example via the data/power bus 101 adapted to transport the output signals from the command means 110 to the drive means 120.

The wheel-equipped apparatus 100 may comprise power supply means 130 (Figure la) that supply power to the drive means 120, to the control unit 200, to the sensor means 123, 124, and to the command means 110 when present. The power supply means 130 may comprise for example one or more lithium, nickel/cadmium batteries and so on and may comprise devices adapted to charge such batteries, such as for example inverters or power supplies. The power supply means 130 may, for example, supply power to the drive means 120, to the control unit 200, to the sensor means 123 , 124 and to the command means 110, when present, via the data/power bus 101. The set of said power supply means 130, drive means 120, sensor means 125, 126, said control unit 200 and command means 110, if present, can be operatively connected to each other, for example via the data/power bus 101, forming an autonomous-drive propulsion system of the wheel-equipped apparatus 100, like for example schematically shown in Figure la. For example, the data/power bus 101 may comprise a digital interface such as for example CANBUS, RS485, and so on or it may comprise an analogue interface.

In accordance with the present embodiment of the invention, a management unit 150 is adapted to manage the autonomous drive of at least one wheel-equipped apparatus 100. The management unit 150 may be used by the user, for example, to set a path of said at least one wheel-equipped apparatus 100 and to display said path to the user himself.

The management unit 150 may comprise, for example, a memory 151, an interface module 152, an input/output module 153, and a processor 154 operatively connected to each other; the management unit 150 may be, for example, a computer, a smartphone, a tablet, and so on.

The memory 151 of the management unit 150 is adapted to store inside it information relating to the autonomous drive of at least one wheel-equipped apparatus 100. Such information may, for example, comprise the data coming from the control unit 200, such as for example the values of position, velocity, operation status of the wheel-equipped apparatus 100, maps of at least a portion of the environment in which the wheel-equipped apparatus 100 operates, and so on.

The information is transmitted and/or received by the management unit 150 in communication with the control unit 200 via the interface module 152 which may be for example a USB, ETHERNET, Wi-Fi, Bluetooth, GSM interface and so on. For example, in the present embodiment of the invention, the control unit 200 may be connected to the interface module 152 of the management unit 150 via a Bluetooth interface.

The input/output module 153 allows the user to interact with the management unit 150. The input/output module 153 may comprise output and input means, such as for example a display and an alphanumeric keypad respectively, or alternatively a touchscreen display in which an alphanumeric keypad and interactive symbols are displayed.

The processor 154 of the management unit 150 is adapted to process the information contained in the memory 151 of the management unit 150, for example in such a way as to generate one or more paths for the wheel-equipped apparatus 100. The processor 154 of the management unit 150 is adapted to display by means of said input/output module 153 the maps generated and/or the operating status of the control unit 200.

The management unit 150 may for example be implemented through a computer product, comprising portions of software code, loadable into a memory of a smartphone, tablet or of a computer equipped with interface means such as for example the USB, ETHERNET, Wi Fi, Bluetooth, GSM interface and so on.

In one embodiment of the invention, the management unit 150 may be a user smartphone which is connected via the interface module 152 to the control unit 200 of the wheel- equipped apparatus 100, such as for example a wheelchair used by the user himself, wherein the control unit 200 is adapted to autonomously drive the wheel-equipped apparatus 100, also as a function of the information exchanged with the management unit 150.

In a further embodiment of the invention, the control unit 200 is adapted to autonomously drive the wheel-equipped apparatus 100 as a function of the information recorded therein. Figure 2 represents an exemplary block diagram of the control unit 200 for correcting odometry errors during the autonomous drive of the wheel-equipped apparatus 100 of Figure la. The control unit 200 may comprise communication means 230, interface means 220, further sensor means 210, storage means 240, and processing means 250. These may be interconnected via a communication bus 201.

The communication means 230 are adapted to establish a communication channel with at least one management unit 150. The communication means 230 may comprise, for example, a USB, CANBUS, ETHERNET, Wi-Fi, Bluetooth, GSM interface, and so on.

The interface means 220 are adapted to receive and transmit input/output information of the control unit 200. The interface means 220 may comprise, for example, a screen for displaying the path of the wheel-equipped apparatus 100, a microphone for vocally commanding the movement of the wheel-equipped apparatus 100, a speaker for audio communications with a remote operator; the screen, the microphone and the speaker may be for example housed in a control panel of the control unit 200. The control unit 200 may comprise a video camera and/or a barcode/RFID reader, for example in order to determine the initial position in which the wheel-equipped apparatus 100 is located.

The further sensor means 210 are adapted to acquire the values of quantities inherent to said control unit 200. For example, the further sensor means 210 may acquire physical quantities useful for the autonomous drive of the wheel-equipped apparatus 100, such as for example accelerometers, speedometers, and so on. The further sensor means 210 may comprise, for example, an inertial measurement unit or inertial IMU platform and encoders operatively connected to the drive wheels 121, 122. The further sensor means 210 may also comprise one or more radars or infrared sensors in such a way as to obtain information useful for avoiding collisions, for example by performing a vertical scan in such a way as to detect the presence of obstacles and/or differences in height (steps) located along a path of the wheel- equipped apparatus 100.

The storage means 240 are adapted to store the information and the instructions of the control unit 200 for correcting odometry errors during the autonomous drive of the wheel-equipped apparatus 100, according to the present embodiment of the invention, and may comprise for example a flash-type solid state memory.

The information may comprise at least one signal representative of mechanical vibrations of at least one drive wheel 121, 122, acquired by the sensor means 125,126, and a set of values and/or parameters useful for the autonomous drive of the wheel-equipped apparatus 100, such as for example a set of maps for the autonomous drive of the wheel-equipped apparatus 100, the status of the inputs and of the outputs of the interface means 220, values of various physical quantities acquired by the further sensor means 210, such as for example values of temperature, electrical current, electrical voltage, and so on. The instructions stored in the storage means 240 will be described in detail below, with reference to the flowchart of Figure 3.

The processing means 250 are adapted to process the information and the instructions stored in the storage means 240, with reference to the communication means 230, to the interface means 220 and may comprise for example an ARM multicore type processor, an Arduino type microcontroller and so on. The processing means 250 perform low-level operations such as for example Path-Finding, Real-Time-Obstacle-Avoidance, and Tip-Over- Prevention operations in Safety-Critical mode, according to the reference standards. The processing means 250 may establish a communication between the control unit 200 and the management unit 150, using the communication means 230. The processing means 250 can process the information and the instructions stored in the storage means 240, with reference to the communication means 230 and to the interface means 220 and can perform high-level operations such as for example Off-Line-Obstacle-Avoidance operations, based on one or more static maps stored in said storage means 240. The processing means 250 may implement telecommunication functions, via the communication means 230, for example with a remote server, with an elevator, and with other home automation devices. The processing means 250 may implement advanced functions such as for example voice recognition of the user commands acquired by the microphone of the control unit 200.

The communication bus 201 is adapted to interconnect said communication means 230, interface means 220, further sensor means 210 and said storage means 240 to the processing means 250.

With reference to Figure 3, an exemplary method of correction of odometry errors during the autonomous drive of the wheel-equipped apparatus 100 is described. As described above, said wheel-equipped apparatus 100 comprises drive means 120 operatively connected to at least two drive wheels 121, 122, sensor means 125,126 operatively connected to said at least two drive wheels 121, 122 and a control unit 200 operatively connected to the drive means 120 and to the sensor means 125,126.

At step 300, an initialization phase of the control unit 200 is carried out that allows its implementation. For example, in this step the processing means 250 verify the operating status of the drive means 120, of the sensor means 125,126 and of the remaining components of the wheel-equipped apparatus 100, for example the power supply means 130 and the command means 110, if present.

At step 310, the control unit 200 determines the initial position of the wheel-equipped apparatus 100. For example, during this phase the control unit 200 can determine this initial position via the interface means 220 comprising an RFID sensor capable of reading one or more tags previously positioned at different locations in an enclosed environment. For example, each tag can be associated with a pair of coordinates specifying a point on a predefined map of the enclosed environment. The wheel-equipped apparatus 100 by positioning itself in the vicinity of the tag can then acquire its initial position with reference to the predefined map, this map having been previously stored in the storage means 240. Additionally or alternatively, the control unit 200 can determine the initial position via the interface means 220 comprising a video camera, or equivalently a camera, capable of reading one or more barcodes, or QR-codes, previously located at different positions in the enclosed environment. For example, each barcode, or QR-code, can be associated with a pair of coordinates specifying a point on the predefined map of the enclosed environment. The wheel-equipped apparatus, by positioning itself near the barcode, or QR-code, and by scanning it, can acquire its initial position with reference to this predefined map. Additionally or alternatively, the control unit 200 can determine this initial position via the interface means 220 comprising the video camera, or equivalently the camera, recognising the enclosed environment by means of image-recognition techniques. For example, the video camera, or the camera, may acquire an image of a part of the enclosed environment in which the wheel-equipped apparatus 100 is positioned, and subsequently the control unit 200 may thus recognise the image of the environment to which the pair of coordinates specifying a point on the predefined map of the enclosed environment is associated. In this way, the wheel-equipped apparatus 100, by performing environmental recognition, can determine its initial position.

Additionally or alternatively, the initial position of the wheel-equipped apparatus 100 can be acquired by the control unit 200 via the interface means 220 comprising, for example, a touchscreen display in which the predefined map is displayed. The user, by touching the touchscreen display of the interface means 220, can select a point in the predefined map corresponding to the initial position of the wheel-equipped apparatus 100. Additionally or alternatively, the user may perform similar operations using the management unit 150 described above, i.e., the user by touching a touchscreen display of the management unit 150 may select a point in the predefined map corresponding to the initial position of the wheel- equipped apparatus 100. This initial position can then be transmitted from the management unit 150 in communication to the control unit 200. Finally, the initial position of the wheel- equipped apparatus 100 may be stored in the storage means 240.

At step 320, the control unit 200 determines a final position of the wheel-equipped apparatus 100. For example, this final position can be represented by a pair of coordinates corresponding to a point in said predefined map. The final position of the wheel-equipped apparatus 100 can be acquired by the control unit 200 via the interface means 220 comprising the touchscreen display in which the predefined map is displayed. The user, by touching the touchscreen display of the interface means 220, can select a point in the predefined map corresponding to the final position of the wheel-equipped apparatus 100. Additionally or alternatively, the user may perform similar operations using the management unit 150, i.e., the user by touching the touchscreen display of the management unit 150 may select a point in the predefined map corresponding to the final position of the wheel-equipped apparatus 100. This position can then be transmitted from the management unit 150 in communication to the control unit 200.

Finally, the final position of the wheel-equipped apparatus 100 may be stored in the storage means 240.

At step 330, the control unit 200 autonomously drives the wheel-equipped apparatus 100 from the initial position to the final position following at least one predefined path of the associated predefined map of the enclosed environment in which the wheel-equipped apparatus 100 moves. The predefined path can be represented as a collection of predefined positions, wherein each predefined position can be represented by means of a pair of coordinates corresponding to a point on said predefined map. For example, during this step, the control unit 200 may generate the control signal of the drive means 120 to perform the autonomous drive of the wheel-equipped apparatus 100.

During this step, the control unit 200 may determine the current position of the wheel- equipped apparatus 100 by means of odometry techniques. For example, the current position can be represented by a pair of coordinates corresponding to a point on said predefined map. The control unit 200 may estimate the current position of the wheel-equipped apparatus 100 as a function of the information received from the further sensor means 210. For example, the control unit 200 may calculate the current position using signals coming from the encoders operatively connected to each drive wheel 121, 122. The signals coming from the encoders may represent the number of revolutions of each drive wheel 121, 122, consequently, the control unit 200 can calculate the current position of the wheel-equipped apparatus 100 on the basis of the number of revolutions and of the geometric dimensions of each drive wheel 121, 122. For example, the control unit 200 can determine the current position by calculating a displacement, performed by the wheel-equipped apparatus 100, multiplying the number of revolutions detected by each encoder by a circumference length of the corresponding drive wheel 121, 122 and vectorically summing said displacement to the initial position, and so on for the subsequent current positions. During step 330, the sensor means 125,126 acquire said at least one signal representative of mechanical vibrations of at least one drive wheel 121, 122. For example, in the present embodiment of the invention, the first sensor means 125 may acquire a first signal 01, representative of the mechanical vibrations of the first drive wheel 121, while the second sensor means 126 may acquire a second signal Q2, representative of the mechanical vibrations of the second drive wheel 122.

The first signal 01, representative of the mechanical vibrations of the first drive wheel 121 can be of analogue or digital type at the output from the first sensor means 125, which may be transmitted to the control unit 200 via the data/power bus 101. Similarly, the second signal 02, representative of the mechanical vibrations of the second drive wheel 122 can be of analogue or digital type at the output from the second sensor means 126, which may be transmitted to the control unit 200 via the data/power bus 101.

Accordingly, the first signal 01 and the second signal 02 allow determining, respectively, the rolling modes of the first drive wheel 121 and of the second drive wheel 122 in contact with the excessively smooth or slippery rest surface 190 on which the wheel-equipped apparatus 100 moves.

For example, the first signal 01 may be represented by a first waveform that may indicate the rolling modes of the first drive wheel 121, likewise the second signal 02 may be represented by a second waveform that may indicate the rolling modes of the second drive wheel 122. By analysing the first waveform and/or the second waveform, it is possible to discriminate between the rolling modes of at least one drive wheel 121,122 deriving from the slippage of at least one drive wheel 121,122, in contact with the rest surface 190, and the rolling modes of at least one drive wheel 121,122, in contact with the rest surface 190, deriving from operating conditions without slippages. For example, a slippage of at least one drive wheel 121,122 may be determined in the case where the first waveform and/or the second waveform correspond to at least one predefined waveform and/or in the case where an amplitude of the first waveform and/or of the second waveform exceeds a predefined threshold value for a predefined time interval.

This advantageously makes it possible to detect the mechanical vibrations due to a possible slippage of at least one drive wheel 121,122 in contact with the excessively smooth or slippery rest surface 190 on which the wheel-equipped apparatus 100 moves. For example, in the case where the first drive wheel 121 is slipping on the rest surface 190, the mechanical vibrations due to the slippage of the first drive wheel 121 may be advantageously captured by the sensor means 125, such as for example a microphone. Accordingly, the first signal 01, representative of the mechanical vibrations of the first drive wheel 121, may be sent to the control unit 200 in order to correct the undesirable effects of the slippage of the first drive wheel 121, for example on the basis of a duration time of the first signal 01, and of a number of rotations performed by the first drive wheel 121 in the duration time of the first signal 01.

During step 330, the control unit 200 generates a corrective signal that controls the drive means 120 in such a way as to independently operate the two drive wheels 121, 122, on the basis of said at least one signal representative of mechanical vibrations of at least one drive wheel 121, 122. For example, in the present embodiment of the invention, the control unit 200 may generate the corrective signal on the basis of the first signal 01, and/or of the second signal Q2. The corrective signal may be represented by an analogue or digital signal that may be transmitted to the drive means 120 via the data/power bus 101.

The corrective signal controls the drive means 120 in such a way as to independently operate the two drive wheels 121, 122, generating an angular velocity difference between said at least two drive wheels 121, 122 on the basis of said at least one signal representative of mechanical vibrations of at least one drive wheel 121, 122. For example, in the present embodiment of the invention, the control unit 200 may generate the angular velocity difference on the basis of the first signal 01, and/or of the second signal Q2.

In particular, the first drive wheel 121 may rotate with a first angular velocity, independently of the second drive wheel 122, which may rotate with a second angular velocity. This results in the generation of the angular velocity difference between the two drive wheels 121, 122. Consequently, the angular velocity difference results in a rotation, even only partial, of the wheel-equipped apparatus 100 which advantageously allows to change the current position of the wheel-equipped apparatus 100 itself. In this way, the wheel-equipped apparatus 100 can roto-translate in such a way as to advantageously correct its direction of movement.

The control unit 200 can determine this torque difference so as to minimize a differential value between the current position and the predefined position of the wheel-equipped apparatus 100, i.e., so as to minimize the odometry errors relative to the position of the wheel-equipped apparatus 100 deriving from the odometry techniques discussed above. This allows the control unit 200 to autonomously drive the wheel-equipped apparatus 100 from the initial position to the final position advantageously minimizing the odometry errors relative to the predefined path.

To this end, the control unit 200 may determine the angular velocity difference between the first drive wheel 121 and the second drive wheel 122 in accordance with a predefined function having as arguments said at least one signal representative of mechanical vibrations of at least one drive wheel 121, 122, such as for example the first signal 01, and/or the second signal Q2. Furthermore, the predefined function may have as arguments one or more values deriving from the further sensor means 210, such as for example the values of the encoders operatively connected to the drive wheels 121 and 122. This predefined function can be determined in accordance with one or more of the following known algorithms: Kalmar filters, Linear-Prediction and Proportional-Integral-Derivative control, PID.

Additionally or alternatively, the control unit 200 may determine the angular velocity difference between the first drive wheel 121 and the second drive wheel 122 in accordance with an output signal of a previously trained neural network having as input signal said at least one signal representative of mechanical vibrations of at least one drive wheel 121, 122, such as for example the first signal 01, and/or the second signal Q2. Furthermore, the neural network may have as input signal one or more values deriving from the further sensor means 210, such as for example the values of the encoders operatively connected to the drive wheels 121 and 122.

The training of the neural network may involve the continuous recording of the signal representative of mechanical vibrations of at least one drive wheel 121, 122 and of the difference between the current position and the predefined position of the wheel-equipped apparatus 100, the predefined position being obtained with Motion-Capture systems, such as for example the Vicon type system (htps : //www. v icon during the training of the neural network itself.

At step 340, the control unit 200 checks whether the wheel-equipped apparatus 100 has reached the final position, e.g. within a predefined error margin. If the final position has not been reached, the control unit executes step 330, otherwise it executes step 350.

At step 350, the control unit 200 performs a termination phase in which all the operations necessary to terminate the autonomous drive of the wheel-equipped apparatus 100 are performed. During this step, the control unit 200 may signal the inoperative status of the wheel-equipped apparatus 100, for example by means of luminous and/or acoustic indicators, such as LED indicators and/or buzzers included in the wheel-equipped apparatus 100 itself.

The advantages of the present invention are therefore evident from the description made. The method and the relative control unit, of correction of odometry errors during the autonomous drive of a wheel-equipped apparatus, advantageously allow to effectively move, according to a predefined path, the wheel-equipped apparatus in enclosed environments without the aid of geolocation systems.

Advantageously, the method and the relative control unit, according to the present invention, make it possible to detect in real time the mechanical vibrations due to a possible slippage of at least one drive wheel of the wheel-equipped apparatus in contact with the excessively smooth or slippery rest surface, on which the wheel-equipped apparatus 100 itself moves. In this way, the method and the relative control unit, according to the present invention, advantageously make it possible to correct the odometry errors during the autonomous drive of the wheel-equipped apparatus.

Naturally, the principle of the invention remaining the same, the embodiments and details of construction can be widely varied with respect to what has been described and illustrated purely by way of non-limiting example, without thereby departing from the scope of protection of the present invention defined by the appended claims.