Gradient-Based Trajectory Optimization with Parallelized Differentiable Traffic Simulation

We present a parallelized differentiable traffic simulator based on the Intelligent Driver Model (IDM), a car-following framework that incorporates driver behavior as key variables. Our vehicle simulator efficiently models vehicle motion, generating trajectories that can be supervised to fit real-world data. By leveraging its differentiable nature, IDM parameters are optimized using gradient-based methods. With the capability to simulate up to 2 million vehicles in real time, the system is scalable for large-scale trajectory optimization. We show that we can use the simulator to filter noise in the input trajectories (trajectory filtering), reconstruct dense trajectories from sparse ones (trajectory reconstruction), and predict future trajectories (trajectory prediction), with all generated trajectories adhering to physical laws. We validate our simulator and algorithm on several datasets including NGSIM and Waymo Open Dataset. The code is publicly available at: https://github.com/SonSang/diffidm.

Authors

• Sanghyun Son 0003
• Laura Zheng
• Brian Clipp
• Connor Greenwell
• Sujin Philip
• Ming Lin 0003

Keywords

• Pedestrians
• Filtering
• Roads
• Traffic control
• Prediction algorithms
• Real-time systems
• Reliability
• Robotics and automation
• Trajectory optimization
• Vehicles
• Gradient-based Optimization
• Traffic Simulation
• Real-world Data
• Intelligent Model
• Trajectory Prediction
• Driver Behavior
• Trajectory Reconstruction
• Deep Learning
• Time Step
• Deep Neural Network
• Running Time
• Dimensional Space
• Number Of Steps
• Linear Interpolation
• Autonomous Vehicles
• Position Error
• Traffic Conditions
• Vehicle Position
• Traffic Problem
• Mean Average Precision
• Trajectory Optimization Problem
• Reconstruction Task
• Softplus
• Historical Trajectory
• Vehicle Acceleration
• Missing Rate
• Adam Optimizer
• Spline Interpolation