A Python-based local trajectory planner using multilayer graphs for autonomous race vehicles, returning cost-optimal action sets.
Graph-Based Local Trajectory Planner is a Python-based framework for generating local trajectories for autonomous race vehicles using a multilayer graph approach. It computes globally cost-optimal solutions for driving actions like keeping straight or passing, filtering out infeasible options to ensure safe and efficient planning. The planner has been validated in real-world racing scenarios, achieving speeds over 200kph.
Autonomous vehicle researchers, robotics engineers, and developers working on high-speed motion planning for race vehicles or dynamic driving scenarios.
It offers a flexible, graph-based approach to trajectory planning with proven real-world performance, providing cost-optimal action sets and tools for integration with custom behavior planners, making it ideal for high-speed autonomous racing applications.
Local trajectory planner based on a multilayer graph framework for autonomous race vehicles.
Open-Awesome is built by the community, for the community. Submit a project, suggest an awesome list, or help improve the catalog on GitHub.
Computes cost-optimal trajectories for each driving action like keeping straight or passing, ensuring efficiency in high-speed scenarios as highlighted in the README.
Filters out infeasible action primitives from the returned set, enhancing safety by providing only viable options for planning.
Successfully tested at speeds over 200kph during Roborace, demonstrating proven performance in extreme racing conditions.
Allows action selection via priority lists or custom behavior planners, enabling adaptable decision-making for dynamic scenarios.
Explicitly disclaimed as not safety-validated in the README, making it risky for commercial use without thorough safety assessments.
The multilayer graph approach may be resource-heavy, potentially limiting real-time performance on low-power or non-specialized hardware.
As research-oriented software, it may lack polished documentation, long-term support, or ease of integration compared to commercial tools.