Special Topics on AGV Obstacle Avoidance | Two Types of Design of Obstacle Avoidance by Benewake LiDAR
"Shelf to person" and "Cargo to person" technologies were the main development focus for the warehouse AGV (Automated Guided Vehicle) industry in 2019. Affected by the COVID-19 epidemic in 2020, the industry is now looking into other emerging applications such as medical logistics AGV. However, no matter how the AGV industry evolves, "avoiding obstacles" is always the fundamental basis.

Various methods are used for AGV obstacle avoidance, each with its own advantages and disadvantages. This article introduces two main LiDAR-based obstacle avoidance methods by Benewake AGV: single-point LiDAR AGV obstacle avoidance and 3D solid-state LiDAR AGV obstacle avoidance.

The Benewake TF series single-point LiDAR obstacle avoidance module is a laser ranging technology based on ToF (Time of Flight). It is used to detect obstacles and avoid obstacles at a short distance of 0.1m-12m. The principles and applications are relatively simple. The radar detects the distance of the nearest obstacle to inform the system whether to brake or decelerate. It is a low-cost and reliable solution for a short-range small obstacle detection mobile robot / AGV.
①. High measurement accuracy and stable perception
②. Small field of view and good collimation to avoid false triggering
③. Data refresh rate between 100-1000Hz, sensitive obstacle avoidance
④. Multiple communication interfaces to choose from, simply plug in and use
⑤. Extremely low power consumption, light weight and easy integration
⑥. No false alarms such as echo interference or uneven ground

Benewake CE30 3D area array solid-state LiDAR is another type of obstacle avoidance method, unlike the well-known single-channel and multi-channel rotation radar with a radar installed on the top of the AGV to achieve 360° or specific angle obstacle avoidance, the CE30 emits an elliptical cone beam to detect obstacle information within a fixed line of sight. It is a preferred low-cost solution for fixed path AGV obstacle avoidance. The main features are as follows:
①. Angular resolution of 1° to accurately identify obstacle locations
②. Customized Obstacle Avoidance Mode or Extended Development Obstacle Avoidance Mode
③. 20Hz initial detection frequency, faster response to obstacles
④. Small size, with built-in holes, can be directly embedded into the machine
⑤. No rotating parts inside, reliable and stable structure
⑥. No false alarms such as echo interference or uneven ground, etc.


Customized obstacle avoidance mode is a mode specially developed for AGV obstacle avoidance. As shown in Figure 5, in this mode, the CE30 will screen the most critical obstacle for the AGV, that is, target A in the warning zone, and provide the distance of this target to AGV. Obstacle avoidance areas can be customized according to customer needs.

The Extended Development Obstacle Avoidance Mode is to provide a point cloud mapping of obstacles for the AGV, so as to "draw" the approximate outline of the obstacles. By default, CE30 will output a depth map and signal strength data of each pixel corresponding to it. The depth value of each pixel in the depth map represents the projection distance of the corresponding detection point of the pixel to the front surface of the radar. As shown in Figure 6, the actual environment image (top) taken by an ordinary grayscale camera, the depth map (middle) taken by CE30 at the same location, and the point cloud image drawn from the depth data (bottom, viewed from a slightly higher position than CE30) respectively are shown.
II. Application of LiDAR in AGV and Warehouse
1. Warehouse AGV Obstacle Avoidance
A CE30 LiDAR is installed directly in front of the warehouse AGV to conduct real-time monitoring of obstacles in front of it. It controls AGV deceleration and braking, assisting in its quick access to goods, intelligent handling and other functions.

2. Warehouse Forklift Obstacle Avoidance
The TFmini-Plus is installed directly in front of the fork of the forklift, and the distance information of obstacles in front of it is recognized and fed back to the forklift. After the forklift system processes it, it performs collision avoidance or auxiliary positioning of the pallet position.

3. Obstacle Avoidance in Stereoscopic Warehouse
Two TFmini Plus are installed at the front and rear of the smart trolley, respectively, to monitor real-time obstacles and distance information, and control the trolley to decelerate or use emergency brake, so as to easily access and transport in the warehouse while avoiding obstacles.

If you are interested to test Benewake sensors, please click Contact Us or send email to bw@benewake.com