Recently, Geek+ submitted a paper on a new task planning algorithm and was accepted by ICDE (IEEE International Conference on Data Engineering, one of the top academic conferences in the field of big data and the flagship conference of IEEE (Institute of Electrical and Electronics Engineers).
In a nutshell: Large robotized warehouses are designed to distribute millions of e-commerce orders. A key to automated item distribution is to plan paths for robots, also known as task planning, where each task is to deliver racks with items to pickers to process and then return the rack back.
The algorithm, which has been successfully applied to the RoboShuttle+P40, ensures our large robot fleet systems run as efficiently as possible by adaptively filling racks using reinforcement learning. The algorithm then finds the most efficient path to move the selected racks. The framework adopts a series of efficient optimizations on both time and memory to handle large-scale item throughput.
Prior solutions were unfit for large-scale warehouses due to the inflexibility to handle items arriving at different times and the low efficiency for high throughput.
Learn more by reading the full study here: ICDE 2022 -- Adaptive Task Planning for Large-Scale Robotized Warehouses
