ICP-Based Pallet Tracking for Unloading on Inclined Surfaces by Autonomous Forklifts
New method uses real-time point cloud tracking to prevent pallet damage during tricky unloading.
Deep Dive
Researchers Takuro Kato and Mitsuharu Morisawa developed a control method for autonomous forklifts. It uses the Iterative Closest Point (ICP) algorithm on real-time point cloud data to track a pallet's position and angle relative to the fork. This allows the forklift to align and withdraw its forks parallel to an inclined surface, like a truck bed, preventing the pallet from being dragged and damaged during unloading. The method was validated in simulations and real-world experiments.
Why It Matters
Solves a major warehouse automation challenge, enabling reliable autonomous unloading on uneven surfaces without product damage.