Robot Trajectory Planning Using State Space Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current trajectory planning methods for robots face high calculation costs due to the need to search through a large state space, often resulting in constrained searches and insufficient exploration, especially when controlling torques around joints.
Innovation Solution
A trajectory planning method that uses a search tree to register states as branches, segments the state space into cells, and employs a known-state registration tree to reduce calculation costs by discarding duplicate states and focusing on the next state's distance to the goal, allowing for a more efficient search without constraints.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a predetermined number of branches are derived from the root or a branch and all branches are searched in the state space, then complete search coverage is achieved, but calculation cost becomes enormous
Solution Approach 1:
The patent segments the state space into discrete cells and organizes states into a tree structure with root, intermediate, and leaf nodes. This segmentation allows the system to process and manage states in manageable units, reducing the computational burden of searching the entire state space while maintaining completeness through systematic exploration of all segments.
Solution Approach 2:
The patent performs preliminary actions by pre-defining the tree structure, cell divisions, and state registration mechanisms before the actual search begins. The system pre-establishes evaluation criteria and registration rules, which streamline the subsequent search process and reduce real-time calculation costs while ensuring comprehensive coverage.
2Productivity
If an evaluation function is used to select branches for derivation, then calculation cost is reduced, but search capability becomes constrained
Solution Approach 1:
The patent implements dynamic branch selection by deriving a predetermined number of branches from each node based on evaluation of multiple candidate branches. The system dynamically adjusts which branches are explored by evaluating their potential to reach the goal state, allowing flexible adaptation to different search scenarios while maintaining computational efficiency through controlled derivation ratios.
3Reliability
If the number of derived branches N and tree depth M increase, then search coverage improves, but the number of branches to search becomes NM resulting in enormous calculation cost
Solution Approach 1:
The patent applies partial action by deriving only a predetermined number of branches from each node rather than exploring all possible branches. This selective approach ensures sufficient search coverage by focusing on the most promising paths while avoiding the combinatorial explosion that would result from exhaustive exploration of all NM branches.
Data Source
AI summary
A trajectory planning system obtains a trajectory for controlling a state of an object toward a goal state. The system includes a search tree generating section which registers a state of the object as a root of a search tree in a state space, registers a next state of the object after a lapse of a predetermined time interval obtained through dynamical relationships during the time interval as a branch of the search tree in the state space. The system further includes a known-state registration tree storing section which stores a known-state registration tree and a known-state registration tree generating section which determines a cell to which the next state belongs among a plurality of cells previously prepared by segmenting the state space, determines whether or not a state which belongs to the cell has already been registered as a branch of the known-state registration tree, discards the next state when a state which belongs to the cell has been registered, and registers the next step as a branch of the known-state registration tree when a state which belongs to the cell has not been registered. The system further includes a trajectory generating section which selects a state whose distance to the goal state is minimum among states registered as branches of the known-state registration tree and obtains a trajectory using a sequence of states in a backward direction from the state toward the root of the known-state registration tree.