Weed control with an autonomous work device

1 . A weed control method for an autonomous work device, wherein the autonomous work device comprises

driving means configured to move the autonomous work device in an outdoor environment,

sensor means configured to sense an environment of the autonomous work device and to generate a sensor signal based on the sensed environment, and

a control circuit configured to control the driving means to navigate the autonomous work device in the environment based on the sensor signal, and the method comprising:

obtaining a location of a target plant in the environment, wherein obtaining the location of a target plant in the environment comprises detecting a plant in the environment based on the sensor signal, determining whether the detected plant is the target plant, and computing the location of the target plant,

switching, by the control circuit, from a standard operating mode into a weed suppression mode, and

controlling, by the control circuit, the driving means in the weed suppression mode, to position at least one wheel of the driving means on the computed location of the target plant and to perform a predetermined movement pattern of the autonomous work device while simultaneously maintaining the at least one wheel in an area at the location of the target plant, for exerting mechanical stress on the target plant,

wherein performing the predetermined movement pattern comprises setting a rotational speed of the at least one wheel to zero and moving at least one other wheel with a rotational speed different from zero for a predetermined time or for a predetermined angle.

2 . The weed control method according to claim 1 , wherein

obtaining the location of the target plant in the environment comprises obtaining, via a user interface, an input of a user including the location of the target plant.

3 . The weed control method according to claim 1 , wherein the method comprises

recording the location of the identified target plant in a map stored in a memory.

4 . The weed control method according to claim 3 , wherein the method comprises

switching the autonomous work device into operating in the weed suppression mode at predetermined time intervals,

wherein, in the weed suppression mode, the method comprises

retrieving recorded locations of the target plant in the work area from the map,

moving the autonomous work device to each of the retrieved locations,

performing at each of the retrieved locations, the steps of

positioning at least one wheel of the driving means on the retrieved location of the target plant, performing the predetermined movement pattern of the autonomous work device while simultaneously maintaining the at least one wheel on the location of the target plant.

5 . The weed control method according to claim 1 , wherein the method comprises

receiving, via an interface, data for defining desired plants and for defining weed from at least one of a manufacturer, a service provider and at least one user of the autonomous work device.

6 . The weed control method according to claim 1 , wherein the method comprises

outputting, via a user interface, image information of the detected plant,

receiving, via the user interface, a user input including information whether the detected plant is a desired plant or weed, and

learning, a classifier used in the determining step to determine whether the detected plant is the desired plant or weed based on the received user input.

7 . The weed control method according to claim 1 , wherein the method comprises

receiving a user input defining at least one movement parameter of the predetermined movement pattern for determining an amount of stress exerted by the autonomous work device on the target plant.

8 . The weed control method according to claim 1 , wherein the method comprises

selecting at least one movement parameter of the predetermined movement pattern for determining an amount of stress exerted by the autonomous work device on the target plant based on at least one of a plant species of the target plant, a plant species of plants surrounding the target plant, and a region of a working area in which the target plant is located.

9 . The weed control method according to claim 7 , wherein the method comprises

determining a size of the detected weed and storing the determined size of detected weed in association with the computed location of the detected weed,

monitoring the stored size of the detected weed over a predetermined time interval, and

in case the stored size of the detected weed increases or remains the same size, performing at least one of

amending the at least one movement parameter of the predetermined movement pattern or selecting another predetermined movement pattern for increasing the amount of stress on the detected weed for the associated location, or

outputting, to a user, a notification on the detected weed including at least information on the associated location of the detected weed.

10 . The weed control method according to claim 1 ,

wherein the sensor means comprises at least one camera sensor, and

detecting the plant includes performing image processing of the sensor signal, in particular detecting the plant and determining the target plant using a classification algorithm or a deep neural network.

11 . The weed control method according to claim 1 ,

wherein the rotational speed different from zero corresponds essentially to a rotational speed of the wheel and the at least one other wheel during standard operation of the autonomous work device.

12 . A non-transitory computer readable storage medium storing a program with program-code means for executing the steps according to claim 1 , when the program is executed on a computer or a digital signal processor.

13 . An autonomous work device comprising

driving means configured to move the autonomous work device in an environment,

sensor means configured to sense an environment of the autonomous work device and to generate a sensor signal based on the sensed environment, and

a control circuit configured to control the driving means to navigate the autonomous work device in the environment based on the sensor signal, and

wherein the control circuit is configured to obtain a location of a target plant in the environment by detecting a plant in the environment based on the sensor signal, determining whether the detected plant is the target plant, and computing the location of the target plant, and

the control circuit is further configured to switch from the standard operation mode into a weed suppression mode, and to control the driving means to position at least one wheel of the driving means on the location of the target plant and to perform a predetermined movement pattern of the autonomous work device while simultaneously maintaining the at least one wheel in an area at the location of the target plant, for exerting mechanical stress on the target plant,

wherein performing the predetermined movement pattern comprises setting a rotational speed of the at least one wheel to zero and moving at least one other wheel with a rotational speed different from zero for a predetermined time or for a predetermined angle.

14 . A weed control system, comprising:

the autonomous work device according to claim 13 ; and

an external user device, wherein

the external user device is further configured to determine the location of the target plant based on an input of a user, and to transmit the determined location via a communication unit of the external user device to the autonomous work device.