HOME > NEWS > Robot Safety Topics | Anti-Falling Applications from Benewake

Robot Safety Topics | Anti-Falling Applications from Benewake

2020-05-15

In terms of application scenarios, robots can be divided into two categories. Indoor robots include logistic AGV, robotic vacuum cleaners, service robots, and restaurant robots. Outdoor robots include delivery robots, grid inspection robots, security patrol robots, and outdoor cleaning robots and etc.

Figure 1 Scenarios from robot

Obstacles such as stairs, pits, and walls will block the robots. It is an important part for robots to stop and detour in time to prevent falling. Based on the requirements of anti-collision of all kinds of robots, this article introduces a kind of laser LiDAR distance sensor TFmini-S specially designed for robots, and details its installation methods and characteristics with practical cases.

Figure 2 stairs and walls

Requirements of anti-falling and obstacle avoidance

There are many factors to choose proper anti-falling and obstacle avoidance sensors. It should be considered the surrounding environment, the size, shape, range, position, posture, materials and etc. All these factors may interfere the robot can make timely responses such as braking, deceleration, turning to avoid falling on the steps or depressions.

1. Classify by range: short-range (0-1m), mid-range (1-4m), and long-range (more than 4 m)

2. Classify by obstacles: big or small, move or static, dark or light, materials, etc.

3. Classify by environment: rain, high temperature, strong light, dark, etc.

Figure 3 dark environment with two size of obstacles

二、LiDAR Obstacle Sensors

There are many sensors involved in robot anti-falling and collision avoidance, which are generally known as infrared, ultrasonic, vision. While LiDAR is relatively unknown from public. For instance, Benewake TF series single point ranging & obstacle avoidance LiDAR, it has the advantages of small size, low false alarm rate, high accuracy, high refresh frequency, meeting the requirements of human eye safety, and not easy to be interfered by strong light and other external environment. Most importantly, based on the continuous mass production, the price is no longer the threshold of LiDAR sensor.

Figure 4 TFmini-S LiDAR product

Principle

By emitting a small beam of laser or near-infrared light to the target, the distance is calculated according to the time difference of signal reflected from the object according to TOF (time of flight) time flight principle. The LiDAR will get the relative position of the object and the robot will avoid it in time.

Figure 5 principle of TFmini-S

Table 一、 Review of obstacle sensors


Installation

Considering that most mobile robots are dual-wheeled and the field of view (FoV) of TF Series is small (2°~3.6°), we advise to install two LiDARs on both sides of the vehicle, which can extend the probing area as well as increase safety redundancies. However, only one LiDAR on the front of the vehicle is also a feasible solution, when the size of the vehicle is small.

Figure 6 Installation method,angle β

We also advise to set the height of the LiDAR to 30-50 cm above the ground, with the tilting angle (β: the angle between LiDAR beam and horizontal surface) in between 20-60°. The field of view should not get blocked by the installations. Otherwise, it may cause inaccurate measurements or reduce the valid range.

For instance, when the LiDAR is installed at the height of 42 cm, i.e. AB=42cm as shown in the figure; given that ∠a=30°, according to the trigonometry, BC=42/Sin∠a=84cm

Conclusion: we recommend setting angle∠a to 20°-60° and the height of the LiDAR to 30-50 cm above the ground; Otherwise, it may cause inaccurate measurements or reduce the valid range.

Application Overview

Self-balance robot

Installed at the height of 42cm on the body of robots, two TFmini-S sensors probe the grounds on both left and right sides. The angle between LiDAR beam and horizontal surface is 15° when the robot stays upright. It varies from 10° to 57° when the robot tilts forward and backward. LiDAR measures the distance to the front ground, and sends the readings to MCU. MCU determines whether there is a staircase or a pit after synthesizing LiDAR readings and robot’s own tilting angle.

Epidemic control robot

TFmini-S is installed at the height of 30cm above the ground, maintaining the angle of 16° to horizontal surface. Therefore, it is able to detect obstacles within the range of 1.1 meters towards the front ground. Distance values will increase when there is a staircase. Considering the variation of LiDAR readings, robots are going to slow down till fully stopped. In this way, robots can prevent themselves from falling, and thus avoid economic loss.

Clean robot

TFmini-S is installed at the height of 50 cm, maintaining the tilting angle of 30°, capable of detecting obstacles within 1 meter on the ground. LiDAR readings will increase when there is a staircase. According to the variation of measured distances, robots slow down to stop in order to prevent themselves from falling, and thus avoid economic losses.

Patrol robot

TFmini-S is installed at the height of 15 cm, maintaining the tilting angle of 60°, capable of detecting obstacles on the front ground. LiDAR readings will increase when there is a staircase. Considering the variation of LiDAR distances, robots will slow down to stop in order to prevent themselves from falling, and thus avoid economic losses

Delivery robot

TFmini Plus is installed at the height of 10cm, probing staircases and pits within 50cm on the front. When LiDAR detects a large variation of distance change, robots will decelerate in order to avoid falling and damage.

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

什么网赚稳定收入 山东群英会app 网赚能赚到钱吗 三分快3计划 pk10开奖记录 金旋风网赚 pps网赚广告联盟挣钱 加拿大28 关于网赚知识技术 重庆百变王牌