For almost all industries that need to transport, store or handle physical goods, logistics center is an important part of the supply chain. Now the number of orders completed in these centers, as well as the types and methods of material handling, are increasing worldwide. Autonomous systems are increasingly used to safely and effectively handle the increase in the number and complexity of logistics center performance. These trends further promote the progress of autonomous mobile robots (AMR), which are regarded as an important part of the operation of logistics centers in the future to improve the safety, flexibility, efficiency and productivity of workers.
Here, we discussed AMR in general, emphasizing the technology to achieve the latest generation of AMR and how it will affect the logistics supply chain.
What is AMR and where is it used?
AMR - unlike their predecessor autonomous robots (AR) or autonomous guided vehicles (AGVs) - move independently, rather than being mounted on fixed positions or tracks, with limited tasks, such as robotic assembly arms. The mobility of AMR enables these robots to provide a wider range of services and greater flexibility in industrial environments, because generally, operators or predetermined paths are not required. This flexibility is why AMR is now used in healthcare, agriculture, warehousing and manufacturing businesses around the world.
The sensors and AI/ML functions designed in modern AMR enable these machines to operate without guidance, so that they can even navigate in unknown environments with unexpected obstacles using Synchronous Positioning and Mapping (SLAM). In addition, the rugged design of many of these machines is conducive to their use in hazardous environments, otherwise it will be difficult for human operators to navigate or complete the required tasks.
Anatomy of AMR
In order for AMR to provide these functions, they need a set of airborne AI/ML resources, motor control/drive technology, battery/charging system, sensors and wireless communication capabilities. Each aspect of these technologies must be carefully designed and implemented to achieve optimal efficiency, safety and productivity.
Brain: artificial intelligence/machine learning system
Artificial intelligence/machine learning systems involve the extensive development and potential use of various hardware solutions, including microprocessor units (MCU), digital signal processors (DSP), field programmable gate arrays (FPGA) and graphics processing units (GPU). Before AI/ML technology is considered effective/secure and deployed, designing and implementing basic AI/ML systems requires a completely different set of tools. Typically, multiple of these hardware solutions are used to most effectively process sensor data, communications, and control systems needed to operate mobility.
In general, the required response time, the frame rate of the visual system and the complexity of the task drive the resource requirements of AMR to play a specific level. Another factor in this regard is development time and resources. These two implementation aspects of AI/ML systems can accelerate the resolution of various AMR challenges, such as 3D vision, motion control or security, by using pre packaged hardware and software packages or bundles.
Brawn: motor control and electric motor technology
In order for AMR to be mobile, it needs drive technology, energy storage and energy transmission/charging systems. These systems are likely to be implemented in the form of electric motors, battery storage and electric energy charging, similar to the use of electric vehicles. For the most effective and reliable AMR electric motor system, the most common are brushless DC (BLDC) motors for driving and step motors for controlling functions. This is similar to the technology used in electric tools and electronic machine equipment, such as computer numerical control (CNC) milling machines. These systems need motor drives and complex control systems to achieve reliable operation of their motion and control.
Heart: battery and charging technology
Although tethering is feasible for tracking or fixed robot systems, AMR usually requires self power. This requires an onboard energy storage and charging system, which may need to be connected to the charger. The chemical composition of Li ion battery is the most common high-density energy storage solution. It has the DC fast charging (DCFC) function, which can transfer grid power to AMR and minimize downtime. Lithium ion batteries require specialized charging, discharging, balancing and other battery safety/health electronics. Although the charging of electronic devices may not be included in the AMR itself, it is necessary to install AC to DC inverters and DC/DC power converters to facilitate AMR connection and self charging, and minimize the energy and time waste of round-trip.
Eyes: sensors, cameras, 3D vision
In order for AMR to avoid obstacles and successfully change the route when the environment in the operation area changes, AMR needs a set of sensors to inform the AI/ML algorithm about the situation. These vision systems are necessary for AMR to successfully navigate, ensure safety, and avoid damage to itself, property, or cargo. These sensor systems can be traditional visual spectral cameras, infrared (IR) cameras, ultrasonic ranging sensors, laser radar (LiDAR), infrared proximity sensors, or even 3D vision systems. The resolution, acquisition time and number of sensors of sensor data determine the data throughput that the AI/ML system needs to process. This process usually needs to be completed within a few milliseconds for optimal performance and security.
Ears and mouth: wireless communication system
If multiple AMRs are deployed, wireless communication is critical to the communication between AMRs and facility systems and between AMRs. Common wireless standards for these purposes include Wi Fi ?、 4G LTE、BLUETOOTH ?， And the current 5G. The type of wireless communication system used depends on the amount of data and type of communication that may be required between the AMR and the facility or other AMRs. If you only occasionally update software from a centralized location, 4G LTE may be sufficient. If you want to exchange a large amount of data, but not necessarily time pressed, Wi Fi may be appropriate. The emerging 5G solution is conducive to large-scale machine type communication (mMTC) and ultra reliable low latency communication (URLLC). If time critical data is being exchanged or certainty may be required, they may be the most feasible wireless solution.
conclusion
AMR is the future of many logistics centers and manufacturing tasks, where the cost of employing human operators is too high or unsafe. AMR is a complex machine with a large number of electronic subsystems, which must be selected, designed, programmed and integrated to achieve reliable, safe and efficient AMR.
