PORTFOLIO

AGV Network: Warehouse applications of mobile robots

What can warehouse mobile robots do today?

Is there any job in a warehouse that mobile robots can’t do? It seems that these automated and often autonomous workmates are evolving to take over every difficult and time-consuming task. Let’s have a look at some of the specialized work of automated guided vehicles (AGVs) and other mobile robots in today’s smart warehouses.

Loading and unloading trailers

Traditional counterbalance forklifts and powered pallet trucks are a common sight in most warehouse docking areas. The operators who use them to load and unload goods vehicles not only display great driving skills, precision and speed but respond quickly to problems and new circumstances. How can AGVs match or improve upon human performance in those operations? It’s a big challenge, but it can be done. Some robotics suppliers have developed special counterbalanced AGVs to take on this role.

Responding to unit loads with large differences in size and shape is still perhaps a bit much to ask of these AGVs. To simplify things, it makes sense first of all to standardize and palletize.

Automated navigation in a variety of trailers is also challenging. Laser guidance, aided by reflector targets, allows the AGVs to navigate accurately within the warehouse and find their way to and from the dock. Inside the trailer, however, they switch to advanced ‘natural navigation’ and use the walls as positioning references.

Natural or free navigation systems are those which allow robots to autonomously survey their surroundings, map their environment and calculate their location. AGVs with this kind of intelligence are often referred to as autonomous mobile robots (AMRs). To navigate within trailers and detect pallet positions, AMRs may be equipped with cameras or LiDAR (light detection and ranging) sensors. In addition, sophisticated software solutions enable them to deal with a range of different loading patterns.  

After unloading, the same AGVs might be used to transport loads directly to their storage destinations in the warehouse. In most operations, however, this would be too slow a process, as the trailer needs to be unloaded quickly so the goods vehicle can move on to its next mission. Instead, goods are unloaded to an intermediate buffer area, close to the trailer. From there, they may be moved to their storage locations via AGVs or automated conveyor systems.

Stacking and retrieving in narrow aisles and high bays

To pack the maximum amount of racking capacity into a warehouse, you need to minimize aisle width and maximize racking height. This is achievable with a very narrow aisle (VNA) lift truck, but for the ultimate in efficient use of all that extra racking space it should be a VNA AGV.

VNA AGVs not only increase stocking density but boost a warehouse operation’s productivity with their fast load-handling performance. Guidance is typically by laser navigation, although efficiency can sometimes be increased further if the AGV switches between an inductive wire system within the aisles and a laser system outside the racking.

A reach truck AGV, guided in a similar way, can also enable increases in storage density – although not as much as a VNA AGV. It is, however, a very versatile machine for goods handling and transport tasks both within and outside the aisles.

A key feature is its reach mast, which can be pushed forward, beyond the truck’s normal compact dimensions, to stack and retrieve palletized loads. The mast and load are then pulled back toward the truck for maneuvering in the aisle or other small spaces. Two robust load legs help to distribute the load’s weight and maintain stability.

Picking and sorting

Order picking is often the most labor-intensive operation in a warehouse, and one in which increased efficiency brings big cost savings. Searching for the required goods is potentially time-consuming and complex, and staff may spend hours each day walking or riding between picking locations.

But what if the goods could come to the picker, instead of the picker travelling to the goods? Think of the time and effort it would save. That’s the idea behind goods-to-person robotic picking solutions.

Perhaps the simplest approach to this is where autonomous mobile robots (AMRs), commanded by software, bring the requested stock-keeping units (SKUs) to a picking station. The picker then takes the necessary items from each shelf or container. Voice-guided or light-guided systems might be used to direct the picker quickly to the items he or she needs. When finished, the SKU is returned to its place.

An automated guided cart (AGC), sometimes referred to as an under-ride cart, is ideal for this purpose. The cart drives underneath the selected SKU, lifts it off the ground and follows a path to its required destination. Navigation may be assisted by SLAM (simultaneous localization and mapping) or QR codes.

More futuristically, there are AMRs which carry out the order picker’s previous role – travelling through the aisles, picking the necessary items and taking them to a packing station. Intelligent navigation is aided by sensors, laser scanners and both 2D and 3D cameras. An arm with a vacuum gripper picks up each item and places it into the robot’s onboard storage space. AMRs of this type can also be designed to pick anywhere from the ground up to heights which would be difficult or impossible for humans to reach unaided.

Another method of taking the labor out of order picking is to employ picking assist autonomous mobile robots (PA AMRs). Human order pickers are each given their own distinct area of the warehouse in which to pick. The PA AMRs travel from picker to picker, requesting specific items via an interactive user interface which gives information on what each item is and where it can be found. Goods are picked and placed in the AMR’s onboard storage area. In this way the mobile robot collects everything needed to fulfil an order, or in some cases several orders. Scanners, RFID equipment and environmental sensors are amongst the technologies helping the AMR to navigate.

All of these order picking mobile robots clearly fall into the AMR category, as they operate autonomously, use sophisticated communication and require certain levels of artificial intelligence. Those which actually pick items themselves are especially dependent on machine learning technology.

Inventory counting and monitoring

Although frequently seen as a tedious warehouse chore, inventory tracking is vital. Without accurate and up-to-date information on the number and location of stock items, warehouses may accidentally run out of particular products. Customers are then let down by failure to fulfil their orders in time. By keeping precise counts and records, customer dissatisfaction is avoided and the supply chain keeps flowing. What’s more, good inventory data can help warehouse managers to maintain the perfect balance between too much stock – which is expensive – and the risk of stockouts.

Traditionally, a member of staff would walk through each aisle and check every shelf. Even with the advent of hand-held scanners, this involves a lot of time and labor. Specialized inventory robots offer a much more efficient alternative.

These AMRs regularly travel throughout the warehouse, using a scanner to record the stock on shelves at all levels. When equipped with natural navigation technology (such as SLAM – simultaneous localization and mapping – or LiDAR – light detection and ranging) they can autonomously work out the best path.

If goods are placed on shelves with their labels clearly visible and facing in the right direction, a relatively simple laser-based barcode scanner may be sufficient. A more advanced image-based scanner will read labels in any direction and has a better ability to interpret imperfect barcodes. It can also be used for labels with QR and data matrix codes, which hold more information. A further option is RFID (radio-frequency identification) tags, which can be read very quickly, at longer distances, even if they are out of sight.

As well as ground-based AMRs, there are drones – otherwise known as unmanned aerial vehicles (UAVs) – which autonomously fly throughout warehouses to locate and count inventory. Their advantages include great speed and unlimited operating heights.

Cleaning and disinfecting

Beside the work of handling, storing, managing and retrieving goods, warehouse managers are responsible for maintaining the workplace. An essential element of this is cleanliness. Dust and dirt can give a bad impression to customers and business partners, present a slipping hazard for staff and vehicles, and cause abrasion damage in machinery. Most importantly, a poorly maintained work environment can be damaging to the health and wellbeing of employees – especially through dust-related illnesses. In recent times, the Covid crisis has also highlighted the importance of cleanliness in avoiding disease transmission.

Floor cleaning and scrubbing machines are fundamental to keeping a warehouse clean, but their operation is costly in terms of time and labor. The answer is to deploy floor cleaning robots instead. These AMRs are supplied with detergents and disinfectants to scrub floors and ensure they are hygienically clean. Some may also be fitted with sprayers to disinfect surfaces above ground level.  

To carry out these functions reliably and safely, their equipment combines sensors, cameras and LiDAR with artificial intelligence capabilities. They navigate a course which covers every inch of floor surface, but swerve to avoid people and objects suddenly appearing in their way.

Together with floor cleaners, warehouse businesses may wish to consider using UV (ultraviolet) disinfection mobile robots to sanitize surfaces at all heights. In this case the AMR is fitted with a UV light emitter. UV light is highly effective at killing viruses, bacteria and fungi, and with the right intensity it can do this over a radius of several meters around the robot. Use of natural navigation ensures that the AMR covers all areas.

Automation today and tomorrow

These are just a few examples of how mobile robots are improving efficiency for warehouse businesses and enhancing the quality of working life for their employees. Although some may sound futuristic, they are all available now. As technologies continue to develop, the case for replacing existing equipment and practices with automated alternatives will undoubtedly grow even stronger.

Returning to the original question of whether there is any warehouse job that mobile robots cannot handle, we must conclude that there are not many. Wherever the task is routine, it seems likely that it will eventually be automated. However, despite the amazing achievements made possible by artificial intelligence and machine learning, there will always be a place for that most astounding of intelligent decision-making systems: the human brain.