What is the after-sales process if the robot malfunctions?

Answer : Contact customer service – Work order assignment – On-site investigation – Follow-up visit – Feedback archiving – Troubleshooting

 

How secure is the robot? 

Answer :Our robot communicates with the robot management system using a dedicated frequency band wireless connection, and data transmission is encrypted using SSL to prevent the robot from being hijacked. The robot body is equipped with an emergency shutdown button, which can be triggered by maintenance personnel in case of abnormalities for immediate shutdown. The robot is powered by lithium-ion batteries, and special materials are added to the battery structure to prevent accidents caused by battery impact. Meanwhile, the BMS (Battery Management System) is monitored in real-time at the software level, collecting and monitoring the state of charge of the power battery pack, dynamically monitoring the voltage and temperature of each battery cell, discharge current, and total voltage of the battery pack during the charging process, preventing overcharging or discharging of the battery, and ensuring that each battery in the battery pack reaches a balanced state, thereby improving the utilization of the battery.

 

How is the recognition rate and accuracy of the robot? Can it completely replace manual inspection?

Answer: Recognition rate: For a single server cabinet, it takes approximately 10-25 seconds. For a 500-square-meter data center with 120 server cabinets, it takes approximately 1 hour. Excluding charging time, theoretically, the maximum density allows for 18 inspections per day. However, in practical applications, inspections are typically conducted 5-10 times per day.

 

What functions does the data center inspection robot have?

Answer: The D1 integrates various artificial intelligence technologies such as AI deep learning, autonomous navigation, visual recognition, voice interaction, and facial recognition. It achieves the intelligent and visualized operation and maintenance of data centers by recognizing indicator lights, instruments, switches, LED screens, as well as real-time monitoring of environmental dynamics and real-time inventory of assets.

1. Intelligent Inspection:
With the cooperation of deep learning and machine vision algorithms, the robot can scan, learn, and accumulate the status of indicator lights for each server cabinet. During each inspection, it observes changes in indicator lights, instrument panels, and switches through cameras. When anomalies are detected during the inspection, the robot notifies data center administrators via the inspection management platform through alerts, text messages, emails, etc.

2. Temperature Monitoring:
Real-time temperature monitoring is achieved through infrared cameras, enabling high-temperature warnings and fire detection.

3. Smart Security:
External personnel access identity authentication is achieved through methods such as facial recognition and account verification, optimizing the timeliness and security issues and vulnerabilities of the previous manual audit process.
When external personnel perform internal operations, the robot can follow for real-time video monitoring and create on-the-job recordings. It can also have real-time intercom with the control room, facilitating maintenance personnel in problem-solving and improving efficiency.

4. Asset Inventory:
Automatic inventory of equipment is achieved through methods such as recognizing equipment QR code asset tags. It detects abnormal movement of assets in and out, and the robot is linked with the software platform to issue timely audible and visual alarms and notify maintenance personnel. Additionally, it can generate asset inventory data reports.

 

Can robots be deployed in modular data centers with doors between them?

Answer: Deployment is possible. SuperDimension has a dedicated implementation team and professional solutions for swing doors and double doors. The approach is to transform from manual push-pull mode to electronic automatic doors. Only one door needs to be modified for each individual module data center, with no impact on the normal operation of the data center.

 

If we launch the robot online, what are the deployment requirements?

Answer: 

1. There should be power sockets available on the walls inside the data center, providing standard 220V power supply. The total power consumption of the robot is approximately 180W.

2. A dedicated WiFi network for the robot should be deployed inside the data center. Typically, for a 500-square-meter data center, 1 AC + 8 APs are needed. For military enterprises, WiFi deployment may not be necessary, but the robot will operate in offline mode, which limits certain functionalities such as online video streaming, remote control, and real-time alerting.

3. The distance between server cabinets inside the data center should be greater than 800mm to allow passage for the robot.

4. A dedicated server should be deployed for the robot management system, with the following configuration requirements: 16C16G, hard disk capacity of at least 1TB, running on CentOS operating system, and supporting virtual machine installation.