Safety tips for unmanned cleaning robots

As technology continues to advance, a variety of functional robots are emerging, with the application of open scenarios becoming an important direction in the development of robotics.

However, the variability of open scenarios also brings new challenges - robots need to face unpredictable factors such as pedestrians, vehicles and other obstacles. In this context, the design of safety features becomes crucial.

As a robotics company specialising in open scenarios, KUSA Technology has always put safety at the forefront of its product design and has innovatively introduced the international automotive functional safety standard ISO26262 into the robotics field, based on the core idea that both cars and robots need to be operated safely and reliably when faced with complex environments.

It is well known that the automotive industry has a wealth of experience and mature standards in functional safety, so applying this standard to robots ensures that they can respond correctly when encountering pedestrians, vehicles and other obstacles in open environments, thus avoiding potential risks.

What is ISO 26262?

As the first functional safety standard for mass-produced automotive products, ISO 26262 provides a comprehensive safety lifecycle reference model covering all phases from project management, development and production to operations, service and end-of-life, and is designed to ensure that automotive products meet the necessary Automotive Safety Integrity Levels (ASILs) to safeguard the functional safety of vehicles.

How does this apply to robot design and development?

K is an engineer who is involved in the design and development of an unmanned cleaning robot. Next, let's follow K's workflow and see how he makes the robot both efficient and safe.

👉Step 1 Hazard Analysis and Risk Assessment

Little K first listed the potential risk points of the unmanned cleaning robot in various working environments, such as the impact of the outdoor environment, collision with non-trash objects, data security, etc. He then analysed the likelihood and severity of the consequences of each risk, and accordingly categorised the risks into low, medium, and high, and then finally, based on the results of these evaluations, he identified the necessary safety features and technical specifications.

To deal with the identified risks, K's team uses a range of technical solutions:

◉ Protecting the robot's internal electronics from damage by using materials with high levels of protection against environmental factors such as extreme temperatures, UV exposure and humidity;

◉ Providing full awareness of the environment by integrating multiple sensors (LIDAR, infrared, ultrasonic and vision cameras, etc.) to avoid collisions between the robot and non-trash objects;

◉ Redundant design on key components to reduce the impact of a particular sensor failure on the system and improve reliability and safety;

◉ Implementation of encryption measures to secure communication data;

......

During programme implementation, K's team also verified the effectiveness of the chosen solutions through field testing to ensure that they could meet the established security standards.

👉Step 2 Security Integrity Level Assignment

After completing the initial risk identification and assessment, Jr. K and his team determined the safety integrity levels applicable to the robot based on the severity and likelihood of those risks, drawing on the concept of quantifying functional safety requirements from the ISO 26262 standard.

To determine the safety integrity level for each potential hazard, the team used an assessment model that combined the severity (S), exposure frequency (E), and controllability (C) of the hazard. Based on this model, each hazard was assigned a safety integrity rating from A (lowest) to D (highest).

Safety Integrity Assessment Model

As an example, if a hazard has a severity of S2, an exposure frequency of E2, and a controllability of C3, then it has a safety rating of A.

Based on the determined risk level, K's team takes a series of safety measures:

◉ Use multiple types of sensors to increase the robustness of the system and ensure that sensor signals can be transmitted through multiple hardware paths to reduce the impact of a single point of failure;

◉ Continuously monitoring the robot's own state of health and performing periodic self-tests of the hardware and software to ensure that it is functioning properly;

◉ Divide the software into separate modules for easy management and updating to improve system flexibility and maintainability;

◉ Write documentation for operating and maintaining the unmanned sweeping robot to guide users in the proper use and maintenance of the robot;

......

In addition, K has established a regular review process of the safety mechanisms to ensure their effectiveness and to adjust and upgrade them as needed to continuously improve the robot's level of safety and security.

👉Step3 V model architecture design

During the design and development of the unmanned cleaning robot, K and his team adopted the V Model design architecture, which emphasises the close connection between requirements, design, implementation and verification.

V Model Architecture

First, K's team needed to define the functional and non-functional requirements and ensure that all relevant parties understood and agreed to them;

Second, design the overall architecture of the unmanned cleaning robot based on the requirements document, including the outline design of hardware and software components, and prototype the robot based on the design document.

Finally, conduct overall testing (hardware + software) of the robot to verify that it meets all predefined security objectives and non-functional requirements, and fix software defects based on test feedback to continuously optimise the performance and functionality of the unmanned cleaning robot.

Through the above steps, the unmanned cleaning robot developed by Little K and his team is not only able to complete tasks efficiently, but also able to operate safely and reliably in a variety of environments.

👉Step4 Process Improvement

In order to improve the maturity of the development process of the unmanned cleaning robot, K and his team drew on the Capability Maturity Model Integration (CMMI-DEV) methodology to continually improve their development process, to continually enhance the robot's functionality and safety, and to keep the product competitive in the ever-changing market and technological environment.

Capability Maturity Model

The work of K and his team has been successfully completed, and as a result of this series of efforts, the unmanned cleaning robot Star Hyun® has been officially unveiled.

In addition to referencing ISO 26262, Star Hyun® was designed to follow the international standard ISO 22737, which sets out detailed system requirements, performance requirements and test specifications for Automated Driving System Dedicated Vehicles (ADS-DVs) with a maximum speed of 32 km/h.

At the hardware level, in order to withstand the harsh weather in outdoor scenarios, StarHyun® selects automotive-grade components that have been rigorously tested to ensure reliability and durability, and employs multi-sensor fusion to provide comprehensive environmental sensing capabilities.

At the software level, StarKeun®'s fusion algorithms provide a more comprehensive understanding of the surrounding environment, accurately identifying and predicting the movement behaviour of obstacles, and making decisions based on the data.

At the system level, Star Hyun® is equipped with a remote management platform to monitor the status of the robot in real time. In case of extreme situations, it allows the operator to remotely take over the control of the robot and guide the robot to safely get out of the trap in time to ensure the safety of the operation.

Currently, the unmanned cleaning robot Star Hyun® has landed in 15+ regions, including Shanghai, Jiangsu, Zhejiang, Guangdong, Sichuan , etc., and has never been involved in a safety incident.

Conclusion

With the advancement of technology and the increasing demand of the society, the application scenarios of unmanned cleaning robots will become more and more extensive. Safety is always the core of our concern. Only by ensuring the safety and reliability of unmanned cleaning robots can we truly create a cleaner and safer living environment for people.

KUSA Explorer

Safety tips for unmanned cleaning robots

Top Picks

KUSA News｜Exhibition Spotlight | KUSA Technology Showcases New Product at 2025 Shenzhen (International) Urban Environmental Industry Expo

KUSA News｜KUSA Technology Named to the "Yangtze River Delta High-End Equipment New Quality Leaders List"

KUSA News｜KUSA Technology Wins Third Prize in the 2025 "Innovation in Shanghai" International Innovation and Entrepreneurship Competition Finals