As autonomous mobile robots (AMRs) move into mass production and see wider adoption in logistics, warehousing, manufacturing, healthcare, and consumer service, the demand for reliability has grown significantly. To promote exchange and standardization of industries, the Autonomous Mobile Robot Alliance (AMRA) organized the “AMRA Seminar 2025” during the TAIROS exhibition on August 21. The seminar not only introduced the latest standard of AMR communication protocol but also featured leading industry experts who shared insights into the latest developments in AMR products, applications, and testing services.

Ken Chang, Technical Manager at DEKRA iST, was invited to speak with the topic “From Testing to Prediction: Accelerated Life Testing as a Key Enabler of AMR Component Reliability,” and he shared his practical experience in reliability validation, explaining how systematic testing and lifetime prediction methodologies can help industry partners enhance the durability and overall quality of AMR products, thereby strengthening their market competitiveness.

Ken Chang first pointed out a common dilemma in the practical application of AMRs: 「Functions can be achieved and tasks can be performed, but durability and stability are not satisfied.」, and this phenomenon cause high failure rates and increasing maintenance costs. He explained that such issues are typically caused by the failure of critical components—such as sensors, cameras, motors, and batteries—which are often the key factors affecting the stability of AMRs.
 

Definition of Accelerated Life Test

Accelerated Life Test (ALT) applies environmental or operational stresses that exceed actual usage conditions to accelerate component aging and failure mechanisms. By doing so, statistical models can be used to estimate the lifetime distribution under normal conditions. Highly Accelerated Life Test (HALT), on the other hand, places products in even more extreme environments with aiming of identifying design weakness and potential risk for improvement.

Accelerated Life Test mainly involves stress conditions such as temperature cycling, high temperature, random vibration, shock, damp-heat exposure, and electrical cycling. Following the tests, results are commonly analyzed using Weibull life models to characterize reliability performance.
To ensure meaningful outcomes, three guiding principles should be followed when designing life tests:
  1. Applied stresses must replicate realistic failure mechanisms;
  2. Sample size and grouping should strike a balance between statistical validity and cost efficiency;
  3. Data from surviving samples should be properly included in the analysis to avoid distortion of results.
Through these testing methods and analytical techniques, product lifetimes can be predicted early, and can help customers pull in risk consideration during design and operation stages. This reduces future maintenance and return costs—capturing the core value of accelerated life testing.
 

Case Study: Accelerated Test of AMR Components

Ken Chang shared three practical case studies to illustrate how accelerated life testing can be applied to validate the reliability of key AMR components:
  1. Motor module testing: Conducted under high current and high temperature conditions, while simultaneously measuring coil temperature, current variation, and vibration indicators. The data was then analyzed using a Weibull distribution to estimate the Mean Time Between Failures (MTBF).
  2. Lithium battery module lifetime analysis: Performed charge–discharge cycling under different temperature and discharge rate (C-rate) conditions to establish a correlation model between State of Health (SoH) and cycle count. The Arrhenius model was then used to predict B10 (10% failure) and B50 (50% failure) lifetimes under standard conditions.
  3. Sensor camera module executed HALT and vibration test: Through Highly Accelerated Life Test (HALT) and random vibration simulations, the resonance response and structural weakness of the camera bracket were observed. Based on these findings, design and improvement recommendations were proposed and effectively enhancing vibration resistance and overall reliability.
Ken Chang further emphasized that the most critical step after testing is accurately converting “lifetime data under accelerated conditions” to “lifetime predictions under real-world conditions.” He cautioned that ignoring non-failed samples may result in biased lifetime estimations. Therefore, by comparing and calibrating experimental data with actual failure records, a comprehensive prediction model can be established—spanning from design through to market deployment.
 

Conclusion: Reliability is the key link connecting design and operation

Ken Chang not only shared a wealth of reliability case studies but also provided concrete recommendations and key insights on the future development of the industry. He emphasized: “Reliability should not be a remedy after the fact, but rather the bridge between design and operation.” When test data and market application data are integrated into a systematic cycle, companies can move from “lifetime estimations” toward “continuously predictable operation decisions,” thereby enabling data-driven reliability management and ultimately enhancing the overall competitiveness in AMR industry.
 
To make your entire PROBLEM SOLVED, we provide professional advisory and validation service.
For more information or service, please feel free to email to 📧 sos@dekra-ist.com
Back to the top