Robotics Enabling Fully‐Integrated Logistics Lines for Supermarkets — REFILLS (Project ID: 731590)

Project

REFILLS proposes to work on three scenarios:

  1. Store monitoring
  2. Collaborative shelf refilling
  3. Autonomous shelf refilling.

The first scenario will implement the acquisition of full information on location, stock and availability of goods within the store. This information will improve the backroom pre-sorting process. The second and third scenario are related to the actual “shelf refilling”. They differ in the level of autonomy of the robot, with corresponding required abilities and functionalities:

  • The collaborative shelf refilling scenario will include physical and cognitive help for the clerk, based on closed monitoring of the stacking situation, and will require safety measures when humans (clerks and/or customers) are present, especially during shop opening times.
  • The autonomous shelf refilling scenario will consider as a key result the robustness in robotic shelf filling, given the variety of items, and is proposed for operation during closing time.

It is expected that collaborative and autonomous shelf refilling will coexist in the future because fullautonomy is not expected to be achieved in an economic way for all items, so both cases need to be investigated.

To fulfil the three scenarios, the following Robotic modules are considered:

  • Low-cost passive-mobility Trolley modules for cartons and waste
  • A robotic mobile platform (Carrier module) for autonomous transport and navigation
  • A passive-mobility Scanning + Pointing module endowed with a pan-tilt unit for visual shelf monitoring and visual guidance of the clerk
  • A passive-mobility Handling module endowed with a manipulator arm and grippers, a pan-tilt unit and multiple sensors

The robotic modules are illustrated in the following figure 

To fulfil the three scenarios, the following Robotic modules are considered:

  • Low-cost passive-mobility Trolley modules for cartons and waste
  • A robotic mobile platform (Carrier module) for autonomous transport and navigation
  • A passive-mobility Scanning + Pointing module endowed with a pan-tilt unit for visual shelf monitoring and visual guidance of the clerk
  • A passive-mobility Handling module endowed with a manipulator arm and grippers, a pan-tilt unit and multiple sensors

The robotic modules are illustrated in the following figure

As shown in the figures, a carrier module can also drag a traditional roll cage or a Trolley module (Autonomous transport unit) to the required store locations, or carry a scanning + pointing module (Autonomous scanning unit) for store monitoring. A carrier can also position a handling module (Mobile handling unit) at required locations in the store for collaborative and autonomous shelf refilling. Another key aspect of this solution is that most of the robotic operations can actually be done when the shop is closed, which significantly reduces safety and cycle time requirements.

The possibility to operate during closing hours also makes the solution attractive even if the robots do not reach human equivalent performance by the end of the project.

REFILLS proposes to develop solutions allowing robots to improve logistics processes in a supermarket, revolutionising their current structure. Although the market for on-line ordering is growing, experts agree that there will always be demand for physical supermarkets as they combine the relaxing sensory experience of shopping with the extra service of human advice.

Even while automation is improving the experience of customers at retail shops both in terms of ordering and customer comfort, there is still a lack of automation in the logistics management of the retail store. Most of the logistics costs arise from items handling, items transportation, shelves replenishment and backroom management. Supermarket clerks perform these time-consuming, repetitive, inefficient, monotonous and wearing tasks. A mobile robotic system in close and smart collaboration with humans can ideally perform the tasks described above provided some challenges are solved. The tasks carried out by a supermarket clerk are rather complex to automate. Therefore, robotic solutions should cover a number of intermediate steps ―with the potential for early exploitation and store and customer benefits― up to the desired level of automation to fulfill the logistics needs of the retail market domain. Within this context, REFILLS proposes the development of robotic systems addressing four main in-store logistics processes:

  1. “Shelf monitoring” necessary to keep supermarket shelves stocked and minimize empty or disarranged shelves, which means missed sales
  2. “Store delivery and pre-sorting” of goods
  3. Shelf filling activities of pre-sorted new deliveries, known as “initial shelf filling”
  4. “Refilling from the backroom” of products for which shelf space, at the time of initial shelf filling, was not sufficient.

Within REFILLS, it is expected that effective product pre-sorting and sequencing is achieved with robots, based on information about available items in the store. If the stocks are correctly monitored, pre-sorting will allow sending only relevant products to the shelves for stacking.

This novel “smart shelf refilling” concept with robotic pre-sorting in the backroom, robotic support in the shop, and robotic continuous updates stock inventory is expected to impact in-store logistics.