DATE: 2026/06/25

AMR Autonomous Mobile Robot Design Standards

AMR Autonomous Mobile Robot Design Standards
To design an autonomous mobile robot that complies with current international regulations, R&D engineers must focus on two core standards: ANSI/RIA R15.08 for North American safety requirements, and ISO 3691-4 for global and European industrial vehicle compliance. At the same time, the safety of batteries, chargers, and electrical systems must also meet the requirements specified in UL 3100. In practical development, compliance is not merely a matter of words; it requires the genuine integration of functional security—typically requiring Performance Level d, Category 3/PLD Cat. 3, or SIL2—into the control architecture. This requires dynamic obstacle avoidance through certified safety LiDAR, dual-channel emergency stop circuits, and the use of safety-level controllers. By accurately mapping dynamic safety zones based on load, speed, and braking distance, and rigorously validating software algorithms for safe navigation and risk mitigation, engineers can systematically address design requirements, avoid expensive hardware rework, and successfully pass third-party compliance testing.

The traditional AGV and the current dynamic AMR are really not a dimensional thing. AMR introduces more complex dynamic security requirements. Whether building or deploying robots, understanding how the “AMR autonomous mobile robot design standards” specifically restricts and guides your architecture design is the key to safety and whether you can obtain a green card for market access.

Three Core Pillars Of AMR Design Standards


From AGV design to AMR design, engineers first have to turn a corner: the former fixed route, now is dynamic, no preset path of autonomous navigation. The biggest hurdle in the middle is how to interpret the constraints of these standards on the physical structure, electrical system and control system of the robot.

ANSI/RIA R15.08: The older generation of standards basically treats mobile robots as fool-like automated guided vehicles running on fixed tracks, but R15.08 explicitly recognizes AMRs’ autonomous decision-making capabilities. This standard requires that when AMR deviates from the original path to avoid obstacles, this obstacle-bypassing action must not cause new safety hazards to the surrounding personnel or equipment.

ISO 3691-4: See this standard again. It focuses on the safety integration of driverless industrial vehicles into the actual workplace. It very specifically specifies the clearance safety channels that must be reserved, the requirements for the active detection of personnel, and the verification process that is essential for safety functions.

UL 3100: As for this standard, if the R15.08 standard tube is collision and operational risk, then UL 3100 is staring at the bottom line of electrical safety. It defines extremely stringent test indicators for the robot’s battery management system, charging interface and power distribution system, mainly to prevent catastrophic accidents such as battery thermal runaway and electrical fire.

How To Implement Functional Security


To smoothly comply with these stringent standards, functional Security must be embedded directly into the robot’s control loop. Once a key component involving safety loses its chain, the control system must be able to detect the fault as keenly as radar and initiate the safety stop action with great force.

Role of the security controller


In any robot that meets compliance requirements, controlling the brain is a top priority. If the R&D team wants to draw the board, write the bottom layer and run a set of security controller certifications from scratch, it is almost a “bottomless pit” and will often delay the progress of the project for more than half a year or even a whole year. So the smart thing to do is to build the system around an industrial-grade AMR controller. An off-the-shelf security controller can serve as a core hub to directly process input signals from the dual-channel emergency stop button, encoder, and security scanner, and perform maintenance PLD Cat. 3 or the complex security logic required for SIL2 ratings.

Sensor integration and dual-channel emergency stop


A compliant hardware layout naturally excludes single-node failures and requires redundant architecture. For example, a security-level radar must be connected to the controller through a fail-safe communication protocol to ensure that the monitoring does not drop out for even a second. Similarly, the emergency stop circuit must be designed as a dual channel. Even if one of the contactors is bonded and welded, the second channel must be able to forcibly cut off the power supply to the driver.

Dynamic Safe Zone Division For Dynamic Vehicles


In order to meet both ISO 3691-4 and ANSI/RIA R15.08, AMR absolutely cannot use just one rigid, fixed-size safety shield. The size of the safe area must be dynamically adjusted anytime and anywhere based on the robot’s physical state and operating environment.

Speed and braking distance


The faster the robot is, the longer the braking distance due to inertia will naturally be. The security controller needs to obtain real-time speed feedback from the security encoder and dynamically expand the security alert range of the radar. Once the robot detects an obstacle in this dynamically elongated area, it must immediately initiate controlled deceleration or emergency braking before any physical contact occurs. This sounds reasonable, but in real-world debugging, speed feedback delays often cause false positives, which requires fine-tuning the configuration.

Loads and dynamic changes in application scenarios


Whether or not to carry cargo, and what cargo to carry, will make a world of difference in the physical and dynamic characteristics of the robot.

Jacking robots: For example, in applications involving pile height/jacking robots, the safety area must take into account the offset of the center of gravity when the goods are lifted. The design must ensure that when the lifting mechanism is in a high position, its high-speed turning must be restricted mechanically and in software to prevent the entire vehicle from rolling over.

Laser navigation forklift: For heavy-duty and gear-handling equipment like laser navigation forklifts, the safety area must not only cover the front of the vehicle body, but also include the fork, cargo status, and turning radius at the rear. The standard clearly requires that the fork height, tilt angle and whether it is loaded must be actively monitored to prevent the goods from sliding or the vehicle from becoming unstable during transportation.

How To Translate Abstract Standards Into Code


Hardware compliance is done, and we’re only halfway there. To meet modern design standards, software algorithms responsible for path planning, navigation, and multi-vehicle scheduling must be rigorously validated and correctly configured to reduce operational risks.

Configure security zone and map parameters


During the R&D and debugging stages, good software configuration tools can help you achieve more with less effort. Using specialized configuration software, engineers can accurately build maps of the physical environment, debug safe obstacle avoidance areas, and delineate specific restricted driving areas on the map, allowing robots to actively speed or limit steering angles when entering these areas.

Standardized safety workflow


When faced with complex business scenarios, low-code development kits like SEER Robotics provide engineers with a good framework to strip away security control logic and business application logic. This can ensure the smooth execution of daily tasks such as docking, forking or charging, without interfering with the underlying security protection due to bugs in the upper-level business logic, thereby avoiding accidental system freezes or unreasonably reporting errors.

Dilution of risk at the group level


According to the logic of ISO 3691-4, the deployment safety of mobile robots is actually shared by the robot itself and the factory environment in which it is located. When there are dozens of vehicles running around in a workshop, a dispatch system needs to be introduced to control traffic. Through group control and scheduling, deadlock in narrow channels can be prevented, one-way traffic can be planned, and virtual speed limits can be set in high-risk areas for mixed pedestrian and vehicle traffic. This actually reduces the edge detection pressure of a single robot safety sensor from a macro perspective.

How To Shorten The Time To Market


In the process of consulting various manufacturers, the most typical mistake I have seen is treating security compliance as a “surprise inspection before the final submission of documents”. Once you have built the whole machine and sent it to a third-party agency for testing, you will be told that the hardware circuits do not meet the PLD requirements. At this time, you may have to knock down the entire vehicle’s electrical architecture and chassis structure and redesign it. Not only is this a waste of money, but project delays will also cause you to miss out on market opportunities.

My suggestion is to choose an ecosystem chain with highly integrated hardware and software—from the underlying core security controller to the upper-level configuration software and group control system. For example, using SEER Robotics’ complete set of hardware and software bases, developers can directly use the design framework that they have considered for international compliance standards. This can save a lot of time to explore and calculate probabilities on their own, making it easier for third-party certification bodies to close the loop when evaluating.

Frequently Asked Questions (FAQ)


Q1: What are the main differences between AGV and AMR safety standards?

Traditional AGVs basically follow older standards, such as EN 1525 or ISO 3691-4, which default to the vehicle running on a fixed, predictable path and stopping when it encounters someone. The AMR safety standard represented by ANSI/RIA R15.08 must consider that AMR has the ability to autonomously avoid obstacles and plan paths autonomously. Therefore, the standard hard requirement is that when the robot deviates from the original path and circles the obstacle, this action must never lead the robot into another dangerous new area.

Q2: Why has PLD Category 3 become the standard safety rating in the AMR industry?

According to the risk assessment guidelines of ISO 13849-1 and ISO 3691-4, when an autonomous mobile device weighing tens or even hundreds of kilograms shuttles through a collaborative environment, the system must have an extremely high level of fault diagnosis and fault tolerance. PLD Category 3 means that even if a single component in the safety control system fails, the system must be able to safely guide it to a stop state without causing the entire safety protection function to be directly disabled.


Q3: How exactly do load and speed affect changes in the safety zone?

This is actually a physical problem. The braking distance depends on the mass of the vehicle, speed and friction on the ground. In order to ensure that the robot stops completely before hitting people or objects, the security controller needs to dynamically adjust the protection fan of the radar according to the real-time encoder vehicle speed and the current load state. The faster you run and the heavier you pull, the larger the safety obstacle avoidance fan surface must automatically be placed.

Q4: What is the relationship between the UL 3100 standard and ANSI/RIA R15.08?

Simply put, ANSI/RIA R15.08 is responsible for the operational safety of the tube robot “after running” such as obstacle avoidance and path planning, while UL 3100 is responsible for the electrical and fire safety of the tube robot “physical in nature”. It stipulates hard tests for battery packs, charging piles, power distribution design, etc., in order to prevent your robot from spontaneously combusting or causing electric shock accidents in the factory due to thermal runaway of the battery.


Q5: Do software configuration tools really help substantively pass compliance certification?

Of course there is. With configuration software like RoboShop, engineers don’t have to guess blindly. You can visually check and map the radar safety protection zone in the map to ensure that the distance set in the software and the physically calculated safety braking distance are exactly the same. The data is traceable, and it is easier for certification bodies to recognize your safety closed loop when issuing test reports.


Author:SEER Robotics Technology Expert.

I have focused on industrial mobile robot development, functional Security integration, and the practical implementation of international safety regulations. By working closely with R&D engineering teams, I help translate complex compliance requirements like ANSI/RIA R15.08 and ISO 3691-4 into reliable, physical architectures. My primary goal is to share structured technical insights that simplify the compliance path, helping manufacturers develop safer lifting robots, autonomous forklifts, and integrated control systems with minimal design friction.