Motion Planning Graph Weighting for Dynamic Object Collision Risk
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing motion planning systems struggle to effectively navigate autonomous vehicles or robots in environments with dynamic objects, as they fail to accurately account for the uncertain movements and interactions of these dynamic objects.
Innovation Solution
A motion planner system that performs a collision assessment on a planning graph for a primary agent, taking into account the actions and reactions of other agents in the environment. The system modifies edge weights based on the probability of collision with dynamic objects, using cost functions to reflect both collision probabilities and object-specific costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the motion planning system accounts for uncertain movements and interactions of dynamic objects, then the safety and reliability improve, but the computational complexity and processing time increase
Solution Approach 1:
The system performs preliminary collision assessment on the planning graph before executing the motion plan. By pre-evaluating collision probabilities for each edge in the planning graph and modifying weights in advance, the system prepares safety information ahead of time, reducing real-time computational burden while maintaining high safety standards.
Solution Approach 2:
The system dynamically adjusts edge weights in the planning graph based on collision probabilities with moving objects. The weights are modified in real-time as objects move and their trajectories change, allowing the motion plan to adapt to dynamic environmental conditions while maintaining computational efficiency through structured graph modifications.
2Measurement precision
If the motion planning system performs real-time collision assessment and modifies edge weights, then the path accuracy and collision avoidance improve, but the processing time increases
Solution Approach 1:
The system performs collision assessment and weight modification as preliminary steps before motion plan execution. By calculating collision probabilities and adjusting edge weights in advance rather than during real-time motion, the system achieves high path accuracy while minimizing processing time impact on actual motion execution.
Solution Approach 2:
The system skips detailed collision detection during motion execution by relying on pre-computed edge weights. The planning graph with modified weights guides the motion planner to naturally avoid high-probability collision areas, eliminating the need for continuous real-time collision checks during motion and significantly reducing processing time.
3Reliability
If the system considers reactions of other agents to actions taken by the primary agent, then the reliability and safety improve, but the computational burden increases
Solution Approach 1:
The system performs preliminary assessment of how other agents might react to the primary agent's actions and incorporates these reactions into edge weight modifications. By evaluating potential reactions and collision probabilities in advance, the system accounts for agent interactions without requiring continuous computational resources during motion execution.
Data Source
AI summary
A motion planner of a computer system of a primary agent, e.g., an autonomous vehicle, uses reconfigurable collision detection architecture hardware to perform a collision assessment on a planning graph for the primary agent prior to execution of a motion plan. For edges on the planning graph, which represent transitions in states of the primary agent, the system sets a probability of collision with another agent, e.g., a dynamic object, in the environment based at least in part on the collision assessment. Depending on whether the goal of the primary agent is to avoid or collide with a particular dynamic object in the environment, the system then performs an optimization to identify a path in the resulting planning graph with either a relatively low or relatively high potential of a collision with the particular dynamic object. The system then causes the actuator system of the primary agent to implement a motion plan with the applicable identified path based at least in part on the optimization.