An embodiment of the present invention provides an artificial intelligence (AI) robot for determining a cleaning route using sensor data, comprising: a sensor unit including at least one of an image sensor, a depth sensor or a shock sensor; a cleaning unit including at least one of a suction unit or a mopping unit; a driving unit configured to drive the AI robot; and a processor configured to: acquire the sensor data from the sensor unit, determine a complex area using the acquired sensor data, create a virtual wall for blocking an entry into the determined complex area, determine the cleaning route in consideration of the created virtual wall, and control the cleaning unit and the driving unit based on the determined cleaning route.

An artificial intelligence (AI) robot includes a body for defining an exterior appearance and containing a medicine to be discharged according to a medication schedule, a support, an image capture unit for capturing an image within a traveling zone to create image information, and a controller for discharging the medicine to a user according to the medication schedule, reading image data of the user to determine whether the user has taken the medicine, and reading image data and biometric data of the user after the medicine-taking to determine whether there is abnormality in the user. The AI robot identifies a user and discharges a medicine matched with the user, so as to prevent errors. The AI robot detects a user's reaction after medicine-taking through a sensor, and performs deep learning, etc. to learn the user's reaction, to determine an emergency situation, etc. and cope with a result of the determination.

A robotic work tool system (200) for defining a working area perimeter (105). The robotic work tool system (200) comprises a robotic work tool (100) and a controller (210). The robotic work tool (100) comprises a position unit (175) and a sensor unit (170). The controller (210) is configured to receive, from the sensor unit (170), edge data indicating whether the robotic work tool (100) is located next to a physical edge (430). The controller (210) is further configured to control the robotic work tool (100) to travel along the physical edge (430) while the edge data indicating that the robotic work tool (100) is located next to the physical edge (430) and to receive, from the position unit (175), position data while the robotic work tool (100) is in motion. The controller (210) is configured to determine, based on the edge data and position data, positions representing the physical edge (430) and to define, based on the determined positions, at least a portion of the working area perimeter (105).

A rover or semi-autonomous or autonomous vehicle may use an image classifier to determine a terrain class of regions of an image of the terrain ahead of the rover or vehicle. The regions of the images are used to estimate the slope of the terrain for the different regions. The terrain class and slope are used to predict an amount of slip the rover will experience when traversing the terrain of the different regions. A heuristic mapping for the terrain class may be applied to the predicted slip amount to determine a hazard level for the rover or vehicle traversing the terrain.

Provided are methods for vehicle scenario mining for machine learning methods, which can include determining a set of attributes associated with an untested scenario for which a machine learning model of an autonomous vehicle is to make planned movements. The method includes searching a scenario database for the untested scenario based on the set of attributes. The scenario database includes a plurality of datasets representative of data received from an autonomous vehicle sensor system in which the plurality of datasets is marked with at least one attribute of the set of attributes. The method further includes obtaining the untested scenario from the scenario database for inputting into the machine learning model for training the machine learning model. The machine learning model is configured to make the planned movements for the autonomous vehicle. Systems and computer program products are also provided.

An air cleaner disposed in an indoor space is disclosed. The air cleaner according to an embodiment of the present invention includes a blowing device including a suction port and a discharging port, a fan motor configured to cause air flow, a purification unit installed in the blowing device to clean air, a flow conversion configured to change a flow direction of air discharged from the discharging port, a communication unit configured to communicate with a moving agent moving in the indoor space, and a processor configured to receive feature information collected by the moving agent and associated with a structure of the indoor space, obtain a type of the indoor space by using the feature information, and control an operation of at least one of the fan motor and the flow conversion device by using the type of the indoor space to adjust at least one of an operation mode, a wind direction, and a wind volume.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for predicting how likely it is that a target agent in an environment will yield to another agent when the pair of agents are predicted to have overlapping future paths. In one aspect, a method comprises obtaining a first trajectory prediction specifying a predicted future path for a target agent in an environment; obtaining a second trajectory prediction specifying a predicted future path for another agent in the environment; determining that, at an overlapping region, the predicted future path for the target agent overlaps with the predicted future path for the other agent; and in response: providing as input to a machine learning model respective features for the target agent and the other agent; and obtaining the likelihood score as output from the machine learning model.

A method for dynamic loop closure in a mobile automation apparatus includes: obtaining mapping trajectory data defining a plurality of trajectory segments traversing a facility to be mapped; controlling a locomotive mechanism of the apparatus to traverse a current segment; generating a sequence of keyframes for the current segment using sensor data captured via a navigational sensor of the apparatus; and, for each keyframe: determining an estimated apparatus pose based on the sensor data and a preceding estimated pose corresponding to a preceding keyframe; and, determining a noise metric defining a level of uncertainty associated with the estimated pose relative to the preceding estimated pose; determining, for a selected keyframe, an accumulated noise metric based on the noise metrics for the selected keyframe and each previous keyframe; and when the accumulated noise metric exceeds a threshold, updating the mapping trajectory data to insert a repetition of one of the segments.

An autonomous vehicle safety system may activate to prevent collisions by detecting that a planned trajectory may result in a collision. If the safety system is overly sensitive, it may cause false positive activations, and if the system isn't sensitive enough the collision avoidance system may not activate and prevent a collision, which is unacceptable. It may be impossible or prohibitively difficult to detect false positive activations of a safety system and it is unacceptable to risk a false negative, so tuning the safety system is notoriously difficult. Tuning the safety system may include detecting near-miss events using surrogate metrics, and tuning the safety system to increase or decrease a rate of near-miss events as a stand-in for false positives.

The disclosure relates to a method for mapping a processing area, in particular for determining a processing area, as part of a navigation method for autonomous robot vehicles. According to the disclosure, said method is characterized in that boundary lines between adjoining mapped and unmapped subareas of the processing area that is to be mapped are identified by comparing distances traveled by the robot vehicle during an initial mapping trip within the processing area, mapping of an unmapped subarea adjoining a boundary line is initiated from a point on one of those identified boundary lines during another mapping trip of the robot vehicle into the unmapped subarea, and a map of the processing area is created on the basis of the subareas mapped by the robot vehicle.