Motion Planning Lattice for Dynamic Obstacle Collision Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing motion planning systems struggle to efficiently navigate autonomous vehicles through environments with dynamic obstacles, failing to account for real-time changes and uncertainties in obstacle movements, leading to potential collisions.
Innovation Solution
A motion planning system using a planning lattice with collision assessment, employing simplified geometric representations and fitted functions for trajectory prediction, allowing for real-time collision avoidance and path optimization through reconfigurable hardware architectures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional motion planning systems are used to navigate autonomous vehicles in environments with dynamic obstacles, then the system can provide basic path planning, but the computational complexity increases and real-time response capability deteriorates
Solution Approach 1:
The patent segments the motion planning problem into discrete lattice states and transitions, where the environment is divided into grid cells and motion is represented as sequences of state transitions. This segmentation allows complex continuous motion planning to be transformed into discrete state space search, reducing computational complexity while maintaining collision avoidance capability.
Solution Approach 2:
The patent changes the parameter representation of dynamic obstacles from continuous trajectory models to discrete occupancy probabilities in lattice cells. By representing obstacle positions as probability values in grid cells rather than continuous coordinates, the system reduces computational burden while maintaining reliable collision avoidance through probabilistic safety margins.
2Measurement precision
If detailed collision assessment is performed for all possible paths, then collision detection accuracy improves, but the time required for path planning increases
Solution Approach 1:
The patent applies partial collision assessment by evaluating only the most promising paths identified through heuristic guidance rather than assessing all possible paths exhaustively. The A* algorithm performs collision detection selectively on paths that are likely to be optimal, achieving sufficient detection accuracy without the time cost of complete path space exploration.
Solution Approach 2:
The patent substitutes traditional mechanical collision detection methods with information-based A* search guidance. Instead of performing detailed collision assessments for all paths, the system uses heuristic information to guide the search toward safe paths, replacing exhaustive mechanical verification with intelligent information-directed exploration.
3Reliability
If the motion planning system accounts for uncertainties in dynamic obstacle movements, then collision avoidance reliability improves, but the computational complexity increases
Solution Approach 1:
The patent introduces probability values as an intermediary representation between continuous obstacle motion uncertainty and discrete lattice planning. Instead of directly handling continuous uncertainty in obstacle trajectories, the system converts uncertainties into discrete probability occupancies in lattice cells, serving as a mediator that simplifies computational complexity while preserving collision avoidance reliability.
4Speed
If real-time motion planning is performed to keep up with environmental changes, then the responsiveness to dynamic obstacles improves, but the computational processing time required increases
Solution Approach 1:
The patent implements periodic re-planning where the A* algorithm is executed at discrete time intervals rather than continuously. Motion planning is performed periodically to generate updated paths based on current environmental perceptions, achieving real-time responsiveness without the excessive computational burden of continuous re-planning. This periodic execution balances response speed with processing time requirements.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A motion planner performs motion planning with collision assessment, using a motion planning lattice that represents configuration states of a primary agent (e.g., autonomous vehicle) as nodes and transitions between states as edges. The system may assign cost values to edges, the cost values representing probability or likelihood of collision for the corresponding transition. The cost values may additionally or alternatively represent a severity of collision, for example generated via a parametric function with two or more parameters and one or more weights. A primary agent and/or dynamic obstacles may be represented as respective oriented bounding boxes. Some obstacles (e.g., road markings, edge of road) may be represented as curves. A trajectory of a primary agent and/or dynamic obstacle may be represented by respective sets of fitted polynomial functions, edges on the planning graph, which represent transitions in states of the primary agent, the system sets value representing a probability of collision, and optionally representing a severity of the collision. 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.