Analysis of the installation position of infrared obstacle avoidance sensor for home service robots

When service robots are used in indoor spaces such as home residences, hotel rooms, office spaces, clubs, etc., obstacle avoidance sensors are required. Usually, one or more of the four sensors, infrared, ultrasonic, lidar, and binocular vision, are used for obstacle avoidance operations. This article will mainly analyze the installation location of infrared obstacle avoidance sensors.  

      The basic principle of infrared obstacle avoidance sensor: using the reflective properties of objects. Within a certain range, if there is no obstacle, the infrared rays emitted will gradually weaken and finally disappear as the propagation distance increases. If there is an obstacle, the infrared rays will encounter the obstacle and be reflected to the sensor receiving head. When the sensor detects this signal, it can confirm that there is an obstacle in front and send it to the single-chip microcomputer. The single-chip microcomputer performs a series of processing and analysis to coordinate the two-wheel work of the service robot to complete the obstacle avoidance action.

      The infrared obstacle avoidance sensor is a distance-adjustable obstacle avoidance sensor designed for wheeled robots. It has a pair of infrared transmitting and receiving tubes. The transmitting tube emits infrared rays of a certain frequency. When the detection direction encounters an obstacle (reflective surface), the infrared rays are reflected back and received by the receiving tube. At this time, the indicator light is on. After circuit processing, the signal output interface outputs a digital signal. The detection distance can be adjusted by the potentiometer knob. The effective distance is 2 to 40 cm. The working voltage is 3.3V-5V. Due to the wide working voltage range, it can still work stably when the power supply voltage fluctuates greatly. It is suitable for a variety of microcontrollers, Arduino controllers, and Raspberry Pi. It can sense changes in the surrounding environment when installed on the robot.

A typical infrared barrier sensor

      Working principle of infrared obstacle avoidance sensor: red wire connects to 5V, yellow wire connects to signal, green GND connects to negative power supply or logic ground on microcontroller. If there is no obstacle in front, it will output high level, if there is an obstacle, the output port (yellow) level will change from high level to low level (0). There is a potentiometer on the back to adjust the obstacle detection distance. In the circuit design, you can add a pull resistor of 10K to 5V to the yellow wire at the output end, and then connect it to the microcontroller for detection, which will be more stable. The microcontroller detection can be realized by external hardware interrupts INT0 INT1, etc.

      The infrared obstacle avoidance sensors currently installed on the service robot are installed in three groups: high, medium and low. For example, the most common installation height is 80mm for low position, 480mm for medium position and 780mm for high position. There is no sufficient standard and basis for setting this installation height. In actual applications, there are many cases where the infrared obstacle avoidance sensors cannot detect nearby obstacles because they are installed too high, too low or not dense enough, thus affecting the actions of the service robot and causing damage to the environment and people. This is all caused by inappropriate installation methods of infrared obstacle avoidance sensors. It is urgent to study more suitable deployment plans for infrared obstacle avoidance sensors based on environmental characteristics, especially obstacle characteristics.

During the research and development of the health care and elderly care service robot, Guangzhou Zero Software Technology Co., Ltd. (Robot Zero) conducted a large number of real-life measurements and scenario simulations, and basically summarized the installation height of infrared sensors required for service robots in indoor spaces such as residences. These experiences were also used on the Husky health service robot.

      During our research and development, we sorted out and summarized the heights of common furniture based on the national standards and specifications related to furniture design, especially analyzing furniture protrusions such as table edges, chair surfaces and backrests, coffee table edges, sofa armrests and backrests, and bed edges, because these protruding parts that extend beyond the bottom and protrude on the vertical plane are the most likely to cause obstacles to the movement of service robots. The following is a description of the dimensions of furniture heights:

1. The height dimensions of table furniture can be 700mm, 720mm, 740mm, and 760mm.

2. The seat height of chairs and stools can be in three specifications: 400mm, 420mm, and 440mm.

3. Generally speaking, the height of the bed edge should be 45 cm; the height of the mattress from the ground is measured at 46~50 cm.

4. The general seat height of a sofa is 350-420mm; the armrest height is 560-600mm; the back height of a single sofa is 700-900mm.

5. Coffee table: height 350-500mm.

Other furniture in the room, including cabinets, boxes, bookshelves, etc., are basically consistent in the vertical plane without obvious protrusions, so the distance measured by the service robot's infrared sensor is consistent at any height. As long as an infrared sensor at any height obtains data, it can represent the perception of the obstacle. Therefore, this type of furniture with a flat facade does not need to consider new installation methods for infrared obstacle avoidance sensors.

      Home appliances, especially those that may appear in the vicinity of the service robot's moving path, such as refrigerators, washing machines, water dispensers, etc., also have flat facades, so there is no need to consider new installation methods for infrared obstacle avoidance sensors.

      Since the thickness of the furniture tabletop plate has three main sizes, namely 16mm, 18mm and 25mm, and the thickness of the tempered glass of the coffee table is more than 10mm, usually 12mm. Therefore, I can touch the protruding surface of the furniture within a range of 10mm from the furniture surface (the highest point of the furniture), which can also be an important basis for the installation height of the service robot's infrared obstacle avoidance sensor.

Based on the characteristics of furniture, we have sorted out some heights that are more sensitive to the actions of service robots:

Therefore, from the table above, we can get the low position, middle position and high position of the infrared obstacle avoidance service robot. With the installation height, we need to further design the layout of the infrared obstacle avoidance sensor.

In order to achieve the best obstacle avoidance effect, we use a total of 24 infrared obstacle avoidance sensors, 12 of which are installed on the front, 3 on each of the two sides, a total of 6, and 6 on the back.

      The low-position infrared obstacle avoidance sensors are installed at a height of 80mm, with two on the front, one on each of the two sides, and two on the back.

      The infrared obstacle avoidance sensors in the middle are installed at five different heights, namely 340mm, 390mm, 420mm, 430mm, and 490mm. Five infrared obstacle avoidance sensors are required to be deployed and installed at all five heights on the front, one infrared obstacle avoidance sensor is installed on the side at a height of 420mm; and three infrared obstacle avoidance sensors are installed on the back at heights of 340mm, 390mm, and 420mm.

    The high-position infrared obstacle avoidance sensors are installed at four different heights, namely 690mm, 720mm, 730mm, and 750mm. Four infrared obstacle avoidance sensors are required to be deployed and installed at all four heights on the front, one infrared obstacle avoidance sensor is installed on the side, the height is 720mm; two infrared obstacle avoidance sensors are installed on the back, the heights are 690mm and 730mm respectively.

      In this case, the infrared obstacle avoidance sensor uses an NPN photoelectric switch; the output state is 0, 1, that is, the high level and low level in the digital circuit. When the target is detected, it is a low level output, and the normal state is a high level output. The output plus a pull-up resistor of about 1K can be connected to the IO port of the microcontroller. Since 24 infrared obstacle avoidance sensors are designed, the number of corresponding microcontroller IOs required is 24.

      This installation position and installation method are important improvements over the existing installation methods of infrared obstacle avoidance sensors: the installation position is optimized according to the height of furniture protrusions, and 24 sets of infrared obstacle avoidance sensors are used in three installation areas of low, medium and high, which are installed in front, on both sides and behind the service robot. The infrared obstacle avoidance sensor can always scan the protruding surface of the furniture, so that the service robot can effectively avoid various types of furniture. It is suitable for various service robots such as household service robots, commercial service robots, and food delivery robots.