In addition to Weibo, there is also WeChat
Please pay attention
WeChat public account
Shulou
2025-01-16 Update From: SLTechnology News&Howtos shulou NAV: SLTechnology News&Howtos > Internet Technology >
Share
Shulou(Shulou.com)06/02 Report--
SLAM, as an important breakthrough in robot autonomous positioning and navigation, is attracting attention in the industry. It is the key technology to realize robot autonomous walking, which can help robot realize real-time positioning and map construction. In practical application, how is SLAM technology realized? Let's find out.
The implementation of this technology mainly includes three steps: preprocessing, matching and map fusion:
pretreatment
Preprocessing is to optimize the lidar raw data, eliminate some problematic data, or filter. We all know that robots want to complete positioning and mapping, need to be equipped with lidar to achieve, lidar can obtain information about the environment where it is located, which is usually called point cloud, but it can only reflect a part of the robot's environment.
matching
Matching is a very critical step, which refers to finding the corresponding position of the point cloud data of the current local environment on the established map. The key is that it directly affects the accuracy of SLAM map construction, which is similar to jigsaw puzzles, which is to find similarities in the already assembled images and determine where to put a new puzzle. In the SLAM process, the point clouds collected by the lidar need to be matched and spliced into the original map, as shown in the red part of the figure below:
Without matching, the resulting map can be messy, as shown in the following image:
map fusion
After the matching step is completed, you can directly enter the map fusion. Map fusion is to splice new data from lidar into the original map and finally complete the map update. As shown below, this process is always accompanied by the SLAM process.
Of course, in the actual application process, the world depicted by the sensor will be inaccurate with the actual situation, and the environment in which the robot is located is easy to change, such as suddenly walking into a person or breaking into a kitten. In the face of complex application environments, many probability algorithms are needed, and filtering methods are used for fusion. After the above processes are executed in turn, the grid map we see is finally produced.
A grid map is a division of the environment into a series of grids, where each grid is given a possible value representing the probability that the grid is occupied. This map, which looks no different from what people know about maps, was first proposed by NASA's Alberto Elfes in 1989 and used on the Mars rover, and is essentially a bitmap image, but each "pixel" represents the probability distribution of obstacles in the actual environment.
The above process does not sound complicated, but it is still very difficult to deal with it well. For example, when realizing the robot loop problem, if the matching algorithm is not accurate enough, or there are many disturbances in the real environment, it may occur that after a circle around the environment, a circular walkway that should have been closed is broken.
For example, a normal map should look like the one on the left, but if it is not handled well, it may become like the one on the right.
In the scene of large environment, the loop problem has to be faced, but in reality, even high-precision sensors such as lidar inevitably have some errors. The difficulty of this problem is that it is not easy to discover the initial error until the robot circles the loop, but it is too late to find out that the loop closure problem is difficult to solve. Of course, the problem is not completely solvable, a good commercial SLAM system can solve the loop problem well. Whether the loop problem can be solved well or not has become an index to judge the strength of the system.
The above is a test conducted by Silan Technology staff in the office. The video on the left is a map built based on the open source ROS robot operating system, and the map on the right is based on SLAMWARE. When the robot circled the field, the map constructed by ROS broke down, and the map constructed by SLAMWARE was a perfect closed loop, which perfectly coincided with the design of Silan Technology Office.
Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.
Views: 0
*The comments in the above article only represent the author's personal views and do not represent the views and positions of this website. If you have more insights, please feel free to contribute and share.
Continue with the installation of the previous hadoop.First, install zookooper1. Decompress zookoope
"Every 5-10 years, there's a rare product, a really special, very unusual product that's the most un
© 2024 shulou.com SLNews company. All rights reserved.