Recently, a hamburger brand has been implementing a service that delivers hamburgers to customers using serving robots. In addition, a domestic robotics company is conducting a pilot project to deliver goods to customers using autonomous transport robots, and in state-of-the-art logistics centers recently established, autonomous logistics robots are carrying out sorting and transporting heavy parcels.

These services are emerging as the optimal method for providing contactless services to both customers and field workers in the COVID-19 environment, and are significantly contributing to securing labor and ensuring the safety of workers at various logistics sites.

■ Sensor technology enabling autonomous driving

To enable such services, mechanical engineering elements such as actuators that move the robot are important, but a robot operating system capable of controlling the robot is considered crucial.

Information acquisition is necessary to control robots, and various sensors are being used for this purpose.

A representative sensor is Lidar.

LiDAR sensors are a technology capable of detecting the distance, direction, speed, temperature, material distribution, and concentration characteristics of an object by shining a laser onto it.

Lidar sensors generally utilize the advantages of lasers, which can generate pulse signals with high energy density and short periods, to be used for more precise observation of atmospheric properties and distance measurement.

LiDAR sensors can be classified into time-of-flight (TOF) and phase-shift methods depending on the laser signal modulation method. The TOF method makes it possible to measure distance by emitting a laser pulse signal and measuring the time it takes for reflected pulse signals from objects within the measurement range to arrive at the receiver.

The phase-shift method emits a laser beam that is continuously modulated with a specific frequency and calculates time and distance by measuring the phase change of the signal reflected back from objects within the measurement range.

Image-based sensor sensors utilizing cameras are also used alongside LiDAR. Tesla's autonomous driving system utilizes 2D image sensors based on cameras, excluding LiDAR. Camera-based image-based autonomous driving estimates the robot's position through feature point matching. However, this method has the drawback of requiring the construction of a positioning database using extensive indoor video data for accurate positioning. Additionally, it suffers from the disadvantage that performance decreases if feature points are distorted or insufficient.

■ SLAM, Key to Autonomous Robot Control

Autonomous driving technology is perfected only when the robot can be freely controlled using information collected through sensors. The technology that makes this possible is SALM (Simultaneous Localization And Mapping).

Slam is a map-based technology that essentially explores the surroundings to draw a map and determine one's location.

SLAM technology is fundamentally based on information acquired through sensors such as LiDAR and cameras, and uses the robot's wheel rotation speed to correct position or other auxiliary devices.

Autonomous robots equipped with SLAM technology can independently draw maps and continuously upgrade the map level, enabling them to navigate even narrow paths and reach their destinations accurately using only coordinates.

■ Securing communication networks such as 5G, LTE, and Wi-Fi is essential

Securing communication is essential for controlling autonomous robots. Consequently, telecommunications companies have recently been actively entering the autonomous robot business.

In the past, the range of movement of autonomous robots was limited due to communication limitations, but recently, with reduced communication costs and the spread of 5G, the limitations on the range of movement of autonomous robots are being resolved.

In particular, for autonomous robots, the establishment of 5G mobile communication is essential as there is a lot of long-distance transport such as logistics.

■ Autonomous Robots Must Securing Legal Frameworks

For the widespread adoption of autonomous robots, securing technology and establishing legal frameworks are absolutely necessary.

Currently, acquiring video information by attaching cameras to the exterior of robots for purposes such as generating movement paths and maps is subject to restrictions on the collection and use of identifiable personal information under the Personal Information Protection Act.

In addition, outdoor autonomous robots are classified as 'vehicles' under the current Road Traffic Act, so their passage on sidewalks and crosswalks is restricted.

In addition, it is unclear exactly whether the responsible party should be the robot operator or the robot manufacturer in the event of an accident caused by an autonomous robot.

Accordingly, it is analyzed that for the activation of autonomous robots, the three elements of sensor technology, communication technology, and legal frameworks must be in balance to give momentum to market creation.