A cleaning robot includes a machine body, a perception system, a control system, and a driving system; the perception system includes a laser distance sensor and a camera; the laser distance sensor is located on a top surface of the cleaning robot; and the camera is mounted on the cleaning robot through a mounting bracket, and a field of view of the camera includes a traveling direction of the cleaning robot. The camera apparatus is applied to the cleaning robot, and provides good shockproof performance and good stability. In addition, when a distance between optical axes of two cameras changes, the camera can be replaced and calibrated at any time, facilitating maintenance and repair.

An agricultural machine includes one or more illuminators to illuminate surroundings of the agricultural machine in a traveling direction thereof, and a controller to control self-driving of the agricultural machine while keeping at least one of the one or more illuminators deactivated at nighttime.

A robotic walking assistant includes a wheeled base having a base and one or more position adjustable wheels connected to the base, a body disposed in a vertical direction, positioned on the wheeled base and having a handle, and a control system that receives command instructions. Each of the one or more wheels is slidable with respect to the base between a retracted position and an extended position in a direction that is substantially parallel to a surface where the wheeled base moves. In response to the command instructions, the control system moves the one or more wheels between the retracted positions and the extended positions.

The disclosed technology provides solutions for improving object detection system based on height map data. A process of the disclosed technology can include steps for receiving image data, receiving height map data, the height map data corresponding with a location of the image data, projecting the height map data onto the image data to generate composite image data, and training an object detection model based on the composite image data. In some aspects, the process can further include steps for localizing one or more objects represented by the image data using the object detection model. Systems and machine-readable media are also provided.

An information processing device includes processing circuitry. The processing circuitry obtain target information that indicates at least one of a distance to a target object or a position of the target object. The processing circuitry generate, based on the target information, map information of a space including a plurality of areas, the map information indicating presence or absence of the target object in a first area included in the plurality of areas, and a detailed position of the target object in the first area.

Described herein are embodiments of a method and system that uses a vertical or upward facing imaging sensor to compute vehicle attitude, orientation, or heading and combines the computed vehicle attitude, orientation, or heading with range bearing measurements from an imaging sensor, LiDAR, sonar, etc., to features in the vicinity of the vehicle to compute accurate position and map estimates.

An autonomous robotic system, and method for automatically monitoring the state of shelves in stores, like retail stores or supermarkets are based on a mobile The mobile robot is capable of autonomously navigating the aisles of a store, with the ability to monitor the condition of product shelves. Specifically, the robotic system solves problems associated with the operation of the shelves, mainly with respect to the detection of incorrect or missing price signs, verification of offer signs, detection of product stock, estimation of product layout, and identification of products misplaced or with errors in the spatial extent assigned to the supplier on a shelf.

The disclosure herein generally relates to the field of autonomous navigation, and, more particularly, to a diverse trajectory proposal for autonomous navigation. The embodiment discloses a hierarchical network based diverse trajectory proposal for autonomous navigation. The hierarchical 2-stage neural network architecture maps the perceived surroundings to diverse trajectories in the form of trajectory waypoints, that an autonomous navigation system can choose to navigate/traverse. The first stage of the disclosed hierarchical 2-stage Neural Network architecture is a Trajectory Proposal Network which generates a set of diverse traversable regions in an environment which can be occupied by the autonomous navigation system in the future. The second stage is a Trajectory Sampling network which predicts a fine-grained trajectory/trajectory waypoint over the diverse traversable regions proposed by Trajectory Proposal Network.

A line laser module, including: a module body; a first image capturing assembly, provided at the module body and comprising a first camera, at least one laser emitter and a first image processing module, wherein the at least one laser emitter is provided adjacent to the first camera and configured to emit a line laser with a linear projection toward outside of the module body, the first camera is configured to capture a first environment image containing the line laser, and the first image processing module is configured to acquire obstacle distance information based on the first environment image; and a second image capturing assembly, comprising a second camera and a second image processing module, wherein the second camera is configured to capture a second environment image, and the second image processing module is configured to acquire obstacle type information based on the second environment image.

A computer-implemented method executed by data processing hardware of a robot causes the data processing hardware to perform operations including obtaining a topological map including waypoints and edges. Each edge connects adjacent waypoints. The waypoints and edges represent a navigation route for the robot to follow. Operations include determining, that an edge that connects first and second waypoints is blocked by an obstacle. Operations include generating, using image data and the topological map, one or more alternate waypoints offset from one of the waypoints. For each alternate waypoint, operations include generating an alternate edge connecting the alternate waypoint to a waypoint. Operations include adjusting the navigation route to include at least one alternate waypoint and alternate edge that bypass the obstacle. Operations include navigating the robot from the first waypoint to an alternate waypoint along the alternate edge connecting the alternate waypoint to the first waypoint.