Inertial Measurement Unit (IMU) is the foundation of various mobile systems, including industrial robots, humanoid robots, unmanned aerial vehicles (UAVs), and immersive mixed reality systems. Although each application has different specific requirements for these systems, designers always face a challenge - providing increasingly accurate real-time direction and motion data for applications such as autonomous mobile robots (AMRs).
This article briefly discusses the various unique challenges faced by AMR positioning. Then, introduce Analog Devices' advanced IMUs and demonstrate how to use these IMUs in indoor environments with Global Positioning System (GPS) coverage to address these challenges, while drawing lessons from broader cross domain use.
Why is positioning a challenge for AMR developers?
AMR is crucial for the productivity of smart factories and warehouses, as it helps simplify material flow, reduce waste, and improve utilization. Ensuring accurate positioning of AMR within the facility is the key to success. In specially constructed facilities, the difficulty of locating AMR can be alleviated by carefully placing targets (reference markers) or optimizing layout, but most AMRs are found in traditional facilities. In these facilities, constantly changing lighting, reflective surfaces, and complex geometric shapes combine to make positioning more difficult.
Moreover, the lack of unified infrastructure such as standardized channel widths or predictable ground markings means that robots need to face more complex navigation and mapping tasks.
The nature of the navigation environment presents two main operational challenges. one
Firstly, robots must perform efficient path planning to determine the optimal route through the environment based on current conditions.
Secondly, robots must be able to accurately locate and continuously update their position and direction in real-time during their movement.
In indoor environments without GPS coverage, these two targets must rely entirely on onboard sensing capabilities and computing resources to achieve.
To address these challenges, AMR employs a combination of various forms of sensors. The visual perception system, including cameras, Light Detection and Ranging (LiDAR), and radar, can provide rich environmental data. For example, odometer systems such as wheel encoders and inertial measurement units (IMUs) directly track the motion of robots based on their movement. Although each type of sensor has unique advantages: some are good at long-range detection, while others are good at precise detection, each type also has its own limitations. Through intelligent combination, AMR can achieve the required redundancy and coverage range, thereby maintaining accuracy under unpredictable dynamic conditions.