For many mobile robots , the environment in which they are located is basically unknown. Building a map requires the robot to start from an unknown location in the unknown environment, and locate itself by repeatedly observing the map features (eg, corners, pillars, etc.) during the movement. Position and attitude, and then build the map incrementally according to its own position, so as to achieve the purpose of simultaneous positioning and map construction, which is the common SLAM technology (Simultaneous localization and mapping) in mobile robot navigation. The working environment of the sweeping robot is different from that of the general mobile robot. It works in the indoor home environment and generally does not change in the big pattern, but the local space sometimes produces some changes (such as the movement of the seat, people, pets). The movement, etc.), requires the robot to constantly update the map on the basis of the original map.
The robot environment map construction problem can be explained as follows: From a certain position, a mobile robot should be able to use its sensors to explore the environment, learn environmental information, interpret the scene, and update or construct the map through certain algorithms. At present, there are many ways to represent environmental maps. For robot navigation, the following three requirements should be met:
1, easy to handle the computer
2, easy to add new information and update the map
3, the robot can rely on the map information to complete specific tasks, such as navigation, search, etc.
Commonly used environment maps can be roughly divided into three types: topological maps, geometric maps, and grid maps. Each representation has its own advantages and disadvantages, which are described separately below.
1. Topological Map
Topological maps are abstract maps that maintain the relative positional relationship of points and lines without necessarily maintaining the correct shape and size, distance, and direction of the graph. It is simplified and adjusted to retain only important information. When it comes to topological maps, you may be a bit strange, but there is a kind of topological map. Many people have to deal with it every day, that is, the subway map. The modern subway map originated from the London subway map drawn by Henry Charles Beck in 1931. There is no scale on it. The relative position and distance between the two points do not necessarily correspond to the actual position and distance. Marking the subway route and the stations along the line in different colors, it looks clear and intuitive, and solves many problems caused by the actual geographical proportion. Most of the current subway maps have evolved from this, and Harry Baker is also known.
In the field of robotics, topological maps are a compact and compact representation of the environment. The map contains only two types of elements: node and line segment. The nodes represent important locations such as obstacles, starting points, corners, and charging bases in the environment, and the line segments represent the connection relationship between important points. The complexity of a topological map is determined by the number of important location points in the environment and the connection between important set points. For example, if there are only two key locations in the environment, there are only two nodes and at most one segment in the topology; if there are three key locations, then there are three nodes and at most three segments, and so on. Under each node, there are at most N(N-1)/2 line segments. The topology map represents the environment, ignoring the geometric features in the environment, and the location information is only given a fuzzy description. The robot only needs to know which node to start from, which node to terminate, which line to select, and the position accuracy is not high.
Topological maps are highly abstract, especially when the environment is large and simple. This method represents the environment as a map in a topological sense, with nodes in the graph corresponding to a feature state and location in the environment. If there is a direct connection between the nodes, it is equivalent to the arc of the connected nodes in the figure. The advantages are: (1) facilitates further path and task planning, (2) storage and search space are relatively small, and computational efficiency is high, and (3) many existing mature and efficient search and inference algorithms can be used. The disadvantage is that the construction of the topology map requires obvious node characteristics, has certain limitations on the full coverage path planning, and has limited description information about the environment, and cannot be used to judge whether the traversal has been completed. In the application of intelligent sweeping robots , topological map nodes often use position coordinates as features, and the map itself has a good application in the unit domain link problem in the unit domain segmentation method.
.jpg)
Electronic motors specilized for zebra blinds.
Zebra Blinds Motor,Multiple Limits Zebra Blind Motor,Radio Receiver Zebra Blind Motor,Zigbee Tubular Zebra Blind Motor
GUANGDONG A-OK TECHNOLOGY GRAND DEVELOPMENT CO.,LTD. , https://www.a-okmotor.com