Sampling-Based Tree Planning for Real-Time Robot Path Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing path planning methods for autonomous driving robots require significant computational resources and often generate trajectories that are either inefficient or prone to collisions due to the need to calculate all possible state values, especially in dynamic environments.
Innovation Solution
A sampling-based optimal tree planning method that extracts child nodes within feasible velocity and acceleration regions, calculates costs for state transitions only for non-colliding nodes, and determines an optimal path by minimizing the sum of costs to the goal point, thereby reducing computational load and ensuring real-time path planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all possible state values are calculated to ensure complete path coverage, then path planning completeness is improved, but computational time increases significantly
Solution Approach 1:
The patent applies partial action by calculating only a subset of feasible state values rather than all possible states. The sampling-based approach selects representative states from the continuous state space, performing sufficient calculations to guarantee optimality within the sampled subset while avoiding the computational burden of exhaustive search across all possible states.
Solution Approach 2:
The patent transforms the continuous state parameters (position, velocity, acceleration) into discrete sampled values. By changing the parameter representation from continuous to discrete sampled states, the system reduces the infinite state space to a manageable finite set while maintaining the essential dynamics characteristics through carefully selected sampling points.
2Stability of the object's composition
If the robot considers all feasible velocity and acceleration values, then motion smoothness is improved, but the number of nodes requiring calculation increases
Solution Approach 1:
The patent applies partial action by considering only a sampled subset of velocity and acceleration values rather than all feasible combinations. This selective sampling maintains motion smoothness by including critical boundary and intermediate values while dramatically reducing the total number of state nodes that require computational processing in the path planning algorithm.
3Measurement precision
If traditional path planning methods calculate all state values, then optimal path accuracy is improved, but real-time performance deteriorates
Solution Approach 1:
The patent achieves real-time performance by performing partial calculations on a sampled subset of states rather than exhaustive calculations. The sampling-based optimal tree planning maintains path accuracy by ensuring the sampled states include all critical decision points and boundary conditions, making the partial search sufficient for optimality while enabling real-time computational speed.
Data Source
AI summary
The present invention relates to a sampling based optimal tree planning method, a recording medium storing a program for executing the same, and a computer program stored in the computer-readable recording medium for executing the same, more particularly to, a sampling based optimal tree planning method, a recording medium storing a program for executing the same, and a computer program stored in the computer-readable recording medium for executing the same for enabling real-time path planning by reducing the number of nodes that require calculation by excluding state values corresponding to input values that a mobile robot cannot select from the calculation and sampling some state values of a set of state values feasible for the robot when planning a path in a tree structure.