The General Situation of SLAM

1. About SLAM

The SLAM(SLAM: Simultaneous Localization and Mapping) is a technology that estimate pose and location, mapping in Unknown environment. Using the robot which can shooting landmark around device in unknown environment, device draw a map in 3D space that by use of collected information.

2. History

SLAM is an active field of research within computer vision and new and improved techniques are constantly emerging. ​Research on the SLAM problem began within the robotics community (arguably with the 1986 papers by Smith and Cheeseman), usually with wheeled robots traversing a flat ground plane. Typically this was done by combining sensor readings (such as from a laser scanner or sonar) with information about the control input (e.g. steering angle) and the measured robot state (e.g. counting wheel rotations). This may seem far removed from tracking a handheld camera freely moving in space, but embodies many of the core difficulties of SLAM, such as creating a consistent and accurate map, and making best use of multiple unreliable sources of information. ​

Kalman filter
Algorithm consists of two stages: prediction and update. The prediction phase predicts the value and accuracy of the current state variable. After the value of the current state variable is actually measured, the update phase updates the current state variable to reflect the difference between the predicted measurement and the actual measurement based on the previously estimated state variable.
Particle filter
The purpose of the particle filter is to predict information with only observations that have continuous incoming information.
One of the simulation-based predictive technologies is called the continuous Monte Carlo method.

3. Field Of Applicate

Typically The application is like these pictures.

Unmanned plane

Robot vacuum cleaner

Exploration robot

4. Utilized Camera sensor

In the general SLAM operates using five types of sensors

RGB - D Camera

Recognition of 3D space and geographical features, measurement of RGB colors.

2D LiDAR

Measurement of the distance data of objects around a two-dimensional plane around the sensor

3D LiDAR

Collects and displays distance data about the 3D space around the sensor in a geometrical form

Wheel encoder
Position measurement by sensor’s angular velocity, angular acceleration.

Gyroscope
Measuring of migrate distance of mobile robot by measuring the number of turns of wheel.

REFERENCE

https://www.kudan.io/post/an-introduction-to-simultaneous-localisation-and-mapping