OpenCDA:An Open Cooperative Driving Automation Framework Integrated with Co-Simulation (2107.06260v3)

Abstract: Although Cooperative Driving Automation (CDA) has attracted considerable attention in recent years, there remain numerous open challenges in this field. The gap between existing simulation platforms that mainly concentrate on single-vehicle intelligence and CDA development is one of the critical barriers, as it inhibits researchers from validating and comparing different CDA algorithms conveniently. To this end, we propose OpenCDA, a generalized framework and tool for developing and testing CDA systems. Specifically, OpenCDA is composed of three major components: a co-simulation platform with simulators of different purposes and resolutions, a full-stack cooperative driving system, and a scenario manager. Through the interactions of these three components, our framework offers a straightforward way for researchers to test different CDA algorithms at both levels of traffic and individual autonomy. More importantly, OpenCDA is highly modularized and installed with benchmark algorithms and test cases. Users can conveniently replace any default module with customized algorithms and use other default modules of the CDA platform to perform evaluations of the effectiveness of new functionalities in enhancing the overall CDA performance. An example of platooning implementation is used to illustrate the framework's capability for CDA research. The codes of OpenCDA are available in the https://github.com/ucla-mobility/OpenCDA.

• The paper introduces OpenCDA, a framework that integrates multi-resolution simulators and modular prototypes to enable robust research in cooperative driving automation.
• The paper details a full-stack system and scenario manager that adhere to SAE J3216 standards, allowing seamless algorithm replacement and evaluation.
• The paper demonstrates OpenCDA's potential via vehicle platooning case studies that assess safety, stability, and efficiency using concrete performance metrics.

OpenCDA: An Open Cooperative Driving Automation Framework Integrated with Co-Simulation

The paper introduces OpenCDA, an open-source framework aimed at facilitating research and development in Cooperative Driving Automation (CDA). The framework provides an integrated co-simulation environment, addressing significant gaps in existing simulation platforms that traditionally focus on single-vehicle intelligence rather than cooperative scenarios.

Framework Composition

OpenCDA is comprised of three principal components:

1. Co-Simulation Platform: Integrates simulators with varying resolutions and purposes, including CARLA and SUMO, offering realistic scene rendering and traffic flow simulation capabilities.
2. Full-Stack Prototype System: Includes modules for perception, computation, actuation, and V2X communication, complying with SAE J3216 standards.
3. Scenario Manager: Manages and evaluates testing scenarios, supporting both default and user-defined inputs.

By facilitating interaction between these components, OpenCDA provides a modular environment in which researchers can validate and compare different CDA algorithms, advancing both individual and collective vehicle intelligence.

Key Features

• Modularity: The framework is highly modular, allowing researchers to replace default algorithms with custom implementations seamlessly.
• Integration: Utilizes CARLA for high-quality scene rendering and SUMO for realistic traffic behavior. The integration provides an optimal environment to evaluate individual vehicle performance and collective traffic systems.
• Benchmarks and Connectivity: Equipped with benchmark algorithms and scenarios, OpenCDA supports varied cooperative levels among CAVs.

Case Study: Vehicle Platooning

The paper provides a comprehensive case paper of platooning, showcasing OpenCDA’s capabilities. The paper includes two testing scenarios: maintaining speed and gap within a single lane, and cooperative merging. Both scenarios aim to assess aspects such as:

• Safety: Evaluated through metrics like Time-to-Collision (TTC) and hazard frequency.
• Stability: Examined by analyzing the inter-vehicular time gap and acceleration smoothness.
• Efficiency: Measured in terms of the time required to complete maneuvers and acceleration standard deviations.

The framework demonstrates effective modularity through the ease of replacing algorithms, such as comparing heuristic-based and Genetic Fuzzy System (GFS) methods in platoon joining scenarios.

Implications and Future Directions

OpenCDA provides a significant contribution to the CDA domain by offering a unified platform for testing and developing cooperative automation strategies. Its flexibility and comprehensive simulation capabilities allow researchers to focus on algorithmic development without being hindered by infrastructure limitations.

The utilization of OpenCDA could lead to advancements in traffic efficiency, safety, and energy consumption in Intelligent Transportation Systems. Furthermore, the documented modularity and benchmark settings encourage ongoing developments and community contributions, potentially integrating future technologies such as advanced communication simulators or more sophisticated perception models.

Conclusion

OpenCDA stands as a valuable tool for exploring and iterating on CDA strategies, facilitating both practical implementations and theoretical research. The framework’s robust design and modular nature make it a promising platform for advancing cooperative automation technologies, paving the way for safer and more efficient transportation systems. As an evolving project, OpenCDA invites continuous enhancements from the research community, offering a solid foundation for future advancements in cooperative driving automation.