An autonomous cleaning robot includes a controller configured to execute instructions to perform one or more operations. The one or more operations includes operating a drive system to move the cleaning robot in a forward drive direction along a first obstacle surface with a side surface of the cleaning robot facing the first obstacle surface, then operating the drive system to turn the cleaning robot such that the side surface of the cleaning robot faces a second obstacle surface, then operating the drive system to move the cleaning robot in a rearward drive direction along the second obstacle surface, and then operating the drive system to move the cleaning robot in the forward drive direction along the second obstacle surface.

A method for operating a cleaning system that comprises at least one self-traveling cleaning device that travels in an environment based on an environment map and carries out cleaning activities. The cleaning device accesses a database, in which multiple cleaning activities are stored. A user accesses the database and defines in advance at least one randomly occurring event, depending on the occurrence of which at least one certain cleaning activity is carried out. The user defines an event-dependent activity scenario and the activity scenario is carried out upon the subsequent occurrence of the defined event. At least one cleaning activity is also scheduled time-dependently, and predefined rules determine whether only the event-dependent activity scenario or only the time-dependently scheduled cleaning activity is carried out if the time of an occurrence of a defined event falls short of a predefined minimum time interval.

A machine control system includes an agricultural work machine having an ECU coupled via a system bus to control engine functions, a GPS receiver, data collector, and specialized guidance system including a stored program. The data collector captures agricultural geospatial data including location data for the work machine and data from the ECU, and executes the stored program to: (a) capture geometries of the farm; (b) capture agricultural geospatial data; (c) automatically classify the agricultural geospatial data using the geometries of the farm, into activity/event categories including operational, travel, and ancillary events; (d) aggregate the classified data to create geospatial data events; (e) match the geospatial data events to a model to generate matched events; (f) use the matched events to generate actionable information for the working machine in real time or near real-time; and (g) send operational directives to the agricultural work machine based on the actionable information.

Systems and methods for updating navigational maps based using at least one sensor are provided. In one aspect, a control system for an autonomous vehicle, includes a processor and a computer-readable memory configured to cause the processor to: receive output from at least one sensor located on the autonomous vehicle indicative of a driving environment of the autonomous vehicle, retrieve a navigational map used for driving the autonomous vehicle, and detect one or more inconsistencies between the output of the at least one sensor and the navigational map. The computer-readable memory is further configured to cause the processor to: in response to detecting the one or more inconsistencies, trigger mapping of the driving environment based on the output of the at least one sensor, update the navigational map based on the mapped driving environment, and drive the autonomous vehicle using the updated navigational map.

Systems and techniques for generating a set of connected segments for a device or system to traverse in order to reach every point of the region (a coverage plan). Nodes defining the region to be traversed define a polygon. The polygon is decomposed into a mesh and a graph of the mesh is generated. The graph may be used to determine a longest funneled path which, in turn, may be used to either optimize for a longest path or to divide the polygon for eroding sides. The longest path and/or erosions are used to define a set of segments. The segments are connected, which in some examples is done via an optimization to minimize an amount of time or energy to traverse all segments and connections. The resultant coverage plan is sent to a system configured to receive the plan and traverse the region.

Provided is a method for controlling a robot cleaner including a first operation of identifying that a manual cleaner and the robot cleaner are turned on, a second operation of identifying, by the robot cleaner, a location of the manual cleaner, a third operation of separating cleaning regions for performing cleaning therein from each other, and a fourth operation of starting, by the robot cleaner, cleaning of a corresponding region after the manual cleaner completes cleaning of the corresponding region.

Disclosed in the present invention are a grass-cutting robot and a control method therefor. The grass-cutting robot comprises a travelling apparatus, a motive power apparatus, a detection apparatus and a control apparatus. The travelling apparatus is configured to facilitate travel of the grass-cutting robot on a physical surface in a first direction. The motive power apparatus is configured to drive the travelling apparatus. The detection apparatus is configured to detect an attitude of the grass-cutting robot. The control apparatus is configured to apply a control signal to the grass-cutting robot when the attitude meets a predetermined condition, the control signal causing resistance to arise in the travelling apparatus, and the resistance causing a tendency of at least part of the travelling apparatus to move in the first direction to be hindered. Further disclosed in the present invention is a control method for a grass-cutting robot. The grass-cutting robot and control method therefor according to one or more embodiments of the present invention can improve the precision of grass-cutting robot control, and increase work effectiveness and safety.

An autonomous electric mower for mowing a lawn comprises a frame, drive wheels, cutting deck, computer, a Lidar sensor, at least one color and depth sensing camera. The computer is programmed and operable to: determine the location of the mower; detect obstacles; and to instruct the mower to avoid the obstacles. Advantageously, the system is operable to analyze the data from the multiple sensors and to instruct the mower to continue to safely operate and cut the lawn despite one or more of the sensors being obstructed. Novel route planning methods are also described.

A trajectory for a vehicle can be generated using a lateral offset bias. The vehicle, such as an autonomous vehicle (AV), may be directed to follow reference trajectory for through an environment. The AV may determine a segment associated with the reference trajectory based on curvatures of the reference trajectory, determine a lateral offset bias associated with the segment based at least in part on, for example, one or more of a speed or acceleration of the vehicle, and determine a candidate trajectory for the autonomous vehicle based at least in part on the lateral offset bias. The candidate trajectory may then be used to control the autonomous vehicle.

A method for determining an energy-efficient path of an autonomous device wherein said autonomous device moves over a global grid of cells into which a given operating area has been split, the method being characterized in that determination of said energy-efficient path comprises the steps of: processing of the current cell (201); taking a measurement σ of the processing (202); classifying the measurement σ to be of a particular level Σ (203), taking into account a predefined division, of the measurements results range, into a plurality of measurements levels; storing said classified measurement in a memory of the autonomous device (204) and associating it with the current cell; selecting a reference probability grid (205); updating (207) the probabilities by applying the reference grid (100) to the global grid at its current position such that every cell on the reference grid (100) corresponds unambiguously to one cell on the global grid;

and moving the autonomous device to a next cell of the global grid (208) and setting said next cell as the current cell (201) in order to process the next cell.