LiDAR Robot Navigation
LiDAR robots navigate by using a combination of localization, mapping, as well as path planning. This article will outline the concepts and explain how they work by using an example in which the robot is able to reach a goal within a row of plants.
LiDAR sensors are relatively low power requirements, allowing them to prolong a robot's battery life and decrease the raw data requirement for localization algorithms. This allows for more iterations of SLAM without overheating GPU.
LiDAR Sensors
The sensor is the core of a Lidar system. It emits laser pulses into the surrounding. These light pulses bounce off surrounding objects in different angles, based on their composition. The sensor measures how long it takes for each pulse to return and uses that information to determine distances. Sensors are mounted on rotating platforms, which allows them to scan the surrounding area quickly and at high speeds (10000 samples per second).
LiDAR sensors are classified based on their intended applications in the air or on land. Airborne lidars are often connected to helicopters or an unmanned aerial vehicles (UAV). Terrestrial LiDAR is typically installed on a stationary robot platform.
To accurately measure distances the sensor must be able to determine the exact location of the robot. This information is typically captured using an array of inertial measurement units (IMUs), GPS, and time-keeping electronics. LiDAR systems make use of sensors to calculate the exact location of the sensor in space and time. This information is later used to construct an 3D map of the surroundings.
LiDAR scanners can also be used to detect different types of surface and types of surfaces, which is particularly useful for mapping environments with dense vegetation. When a pulse passes through a forest canopy, it is likely to generate multiple returns. Usually, the first return is attributable to the top of the trees, while the final return is related to the ground surface. If the sensor captures these pulses separately this is known as discrete-return LiDAR.
Discrete return scanning can also be helpful in analysing surface structure. For example, a forest region may yield a series of 1st and 2nd returns with the last one representing bare ground. The ability to separate these returns and record them as a point cloud allows for the creation of detailed terrain models.
Once a 3D model of the environment has been created and the robot has begun to navigate using this information. This involves localization and creating a path to get to a navigation "goal." It also involves dynamic obstacle detection. This is the process that detects new obstacles that were not present in the map's original version and then updates the plan of travel in line with the new obstacles.
SLAM Algorithms
SLAM (simultaneous localization and mapping) is an algorithm that allows your robot to create an outline of its surroundings and then determine where it is in relation to the map. robot with lidar utilize the information for a number of tasks, such as the planning of routes and obstacle detection.
To enable SLAM to work, your robot must have a sensor (e.g. laser or camera), and a computer with the appropriate software to process the data. You'll also require an IMU to provide basic information about your position. The system can determine your robot's location accurately in an unknown environment.
The SLAM process is extremely complex and a variety of back-end solutions exist. No matter which solution you choose to implement the success of SLAM it requires constant communication between the range measurement device and the software that collects data and the robot or vehicle. This is a highly dynamic process that is prone to an unlimited amount of variation.
As the robot moves, it adds scans to its map. The SLAM algorithm will then compare these scans to previous ones using a process called scan matching. This allows loop closures to be identified. The SLAM algorithm updates its estimated robot trajectory when loop closures are discovered.
The fact that the surrounding changes in time is another issue that makes it more difficult for SLAM. For example, if your robot travels through an empty aisle at one point, and is then confronted by pallets at the next location it will be unable to finding these two points on its map. The handling dynamics are crucial in this situation, and they are a feature of many modern Lidar SLAM algorithms.
SLAM systems are extremely efficient at navigation and 3D scanning despite these limitations. It is particularly useful in environments that don't rely on GNSS for positioning, such as an indoor factory floor. It is important to note that even a well-designed SLAM system can be prone to mistakes. It is vital to be able to detect these issues and comprehend how they affect the SLAM process in order to fix them.
Mapping
The mapping function builds a map of the robot's environment that includes the robot as well as its wheels and actuators, and everything else in the area of view. This map is used to aid in localization, route planning and obstacle detection. This is an area in which 3D lidars are extremely helpful because they can be used as an actual 3D camera (with one scan plane).
Map creation can be a lengthy process but it pays off in the end. The ability to create an accurate, complete map of the surrounding area allows it to carry out high-precision navigation, as being able to navigate around obstacles.
In general, the greater the resolution of the sensor, then the more accurate will be the map. Not all robots require maps with high resolution. For example, a floor sweeping robot might not require the same level of detail as a robotic system for industrial use navigating large factories.
This is why there are a number of different mapping algorithms for use with LiDAR sensors. One of the most popular algorithms is Cartographer which employs the two-phase pose graph optimization technique to correct for drift and maintain a consistent global map. It is particularly effective when used in conjunction with Odometry.
Another option is GraphSLAM that employs a system of linear equations to model constraints of a graph. The constraints are represented by an O matrix, and a X-vector. Each vertice in the O matrix is a distance from a landmark on X-vector. A GraphSLAM update is an array of additions and subtraction operations on these matrix elements which means that all of the X and O vectors are updated to account for new observations of the robot.
Another efficient mapping algorithm is SLAM+, which combines mapping and odometry using an Extended Kalman filter (EKF). The EKF updates not only the uncertainty of the robot's current position, but also the uncertainty of the features mapped by the sensor. This information can be used by the mapping function to improve its own estimation of its position and update the map.
Obstacle Detection
A robot must be able to sense its surroundings to avoid obstacles and reach its final point. It utilizes sensors such as digital cameras, infrared scanners sonar and laser radar to sense its surroundings. It also makes use of an inertial sensors to monitor its position, speed and orientation. These sensors allow it to navigate safely and avoid collisions.
One of the most important aspects of this process is the detection of obstacles that consists of the use of sensors to measure the distance between the robot and the obstacles. The sensor can be attached to the vehicle, the robot, or a pole. It is important to remember that the sensor could be affected by various factors, such as rain, wind, and fog. It is important to calibrate the sensors prior to every use.
The results of the eight neighbor cell clustering algorithm can be used to identify static obstacles. This method is not very precise due to the occlusion created by the distance between the laser lines and the camera's angular velocity. To address this issue, a method called multi-frame fusion has been used to increase the accuracy of detection of static obstacles.
The method of combining roadside unit-based as well as obstacle detection using a vehicle camera has been proven to increase the efficiency of processing data and reserve redundancy for subsequent navigational tasks, like path planning. The result of this method is a high-quality picture of the surrounding area that is more reliable than one frame. The method has been tested with other obstacle detection methods like YOLOv5, VIDAR, and monocular ranging, in outdoor comparison experiments.
The results of the experiment showed that the algorithm was able to correctly identify the height and location of an obstacle, as well as its tilt and rotation. It also had a good performance in detecting the size of an obstacle and its color. The method was also reliable and steady, even when obstacles moved.