Smart mower and smart mowing system
A smart mower includes a camera for collecting image data of the environment around the smart mower; an inertial measurement unit (IMU) for detecting pose data of the smart mower; a memory at least used for storing an application program for controlling the smart mower to work or travel; and a processor for calling the application program, fusing the image data collected by the camera and the pose data acquired by the IMU, performing simultaneous localization and mapping (SLAM) of the smart mower, and generating a navigation or mowing action instruction.
1 . A smart mower, comprising:
a camera for collecting image data of an environment around the smart mower;
an inertial measurement unit (IMU) for detecting a pose data of the smart mower;
a memory storing an application program for controlling the smart mower to work or travel;
a processor for calling the application program, fusing the image data collected by the camera and the pose data acquired by the IMU, performing simultaneous localization and mapping (SLAM) of the smart mower, and generating a navigation or mowing action instruction; and
a global positioning system (GPS) sensor, wherein:
(a) when all of the camera, the IMU, and the GPS sensor fail, traveling is stopped and a maintenance reminder is sent;
(b) when two of the camera, the IMU, or the GPS sensor fail, a remaining one of the camera, the IMU, or the GPS sensor is used for positioning and navigation for a preset time; and
(c) when only one of the camera, the IMU, or the GPS sensor fail, a remaining two of the camera, the IMU, or the GPS sensor are used for positioning and navigation.
2 . The smart mower of claim 1 , further comprising a body and wherein the camera is mounted on a front side of the body.
3 . The smart mower of claim 1 , wherein the application program is capable of distinguishing between grassland and non-grassland according to a feature point of a two-dimensional (2D) plane in the image data and a texture feature of the grassland for comparison and automatically generating a mowing region boundary with a boundary line between the grassland and the non-grassland as discrete anchor points through visual-inertial fusion SLAM.
4 . The smart mower of claim 1 , further comprising a cutting blade, wherein the application program is capable of distinguishing between grassland and non-grassland according to a feature point of a 2D plane in the image data and a texture feature of the grassland for comparison and, when a current workplane is not the grassland, stopping rotating the cutting blade.
5 . The smart mower of claim 1 , wherein the application program is capable of determining a type of a current workplane according to a feature point of a 2D plane in the image data and a texture feature of a common type of ground preset by the application program for comparison and, when the current workplane comprises a plurality of ground types, controlling the smart mower to travel to the ground with greater hardness among the plurality of ground types.
6 . The smart mower of claim 1 , wherein the application program further comprises an object recognition program and the application program is capable of selecting a corresponding obstacle avoidance strategy according to an obstacle category recognized by the object recognition program.
7 . The smart mower of claim 1 , further comprising a light, wherein the application program is capable of acquiring a light intensity of current environment according to the image data and, when the light intensity is less than a first light intensity threshold, turning on the light.
8 . The smart mower of claim 1 , wherein when the GPS sensor fails, the processor fuses the image data collected by the camera and the pose data acquired by the IMU to perform simultaneous localization and mapping (SLAM) of the smart mower; when the camera fails, data of the IMU is used to verify result of the GPS; when the IMU fails, visual simultaneous localization and mapping (vSLAM) is performed, after each frame of image processing, a vSLAM result and a current GPS positioning result are sent to a Kalman filter at the same time.