Robotic Pallet Placement Using Bounded Tree Search Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in efficiently stacking dissimilar items on pallets due to the computational burden of simulating all possible scenarios for item placement, leading to unstable and inefficient palletization.
Innovation Solution
A robotic system using a machine learning-based approach with a bounded tree search and pruning techniques to determine optimal item placements, considering both current and future items, while reducing computational complexity through a scoring function and cost evaluation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Stability of the object's composition
If all possible scenarios for item placement are simulated to ensure stable palletization, then stacking stability is improved, but computational time and resources increase significantly
Solution Approach 1:
The system performs preliminary actions by using machine learning models to predict future item arrivals and pre-calculate optimal placement strategies. The bounded tree search explores only relevant future scenarios rather than all possible placements, and pruning techniques eliminate branches that cannot lead to optimal solutions. This allows the system to prepare placement plans in advance while avoiding exhaustive computation of every possible scenario.
2Productivity
If machine learning models consider future items to optimize current placement, then palletization efficiency is improved, but computational complexity increases
Solution Approach 1:
The system applies partial action by considering only a bounded number of future items (e.g., next N items) rather than all future items indefinitely. The bounded tree search limits the depth of exploration to a predetermined level, and pruning techniques discard branches that exceed computational thresholds. This partial consideration of future scenarios provides sufficient optimization for efficient palletization while keeping computational complexity manageable.
3Productivity
If robotic systems are used to automate pallet stacking, then labor efficiency is improved, but adaptability to varied item types and container configurations decreases
Solution Approach 1:
The system achieves adaptability through parameter changes by using machine learning models that can dynamically adjust to different item attributes (size, weight, shape) and container configurations. The scoring function evaluates multiple placement criteria and can be reweighted based on item characteristics. The bounded tree search adapts its exploration based on item properties, allowing the robotic system to handle varied item types and container types while maintaining automated efficiency.
Data Source
AI summary
A robotic system is disclosed. The system includes a memory that stores a machine learning-based model to provide a scoring function value for a candidate item placement on a pallet on which are plurality of items are to be stacked given a current state value of the pallet and a set of zero or more items placed previously. The system includes one or more processors that use the model to determine a corresponding score for each of a plurality of candidate placements for a next item to be placed and the current state value associated with the current state of the pallet and a set of zero or more items placed previously, select a selected placement based at least in part on the respective scores, control a robotic arm to place the next item according to the selected placement.