1 October 2019 in Environmental Modelling & Software
15 April 2024 in IEEE Robotics and Automation Letters
Autonomous racing is a critical research area for autonomous driving, presenting significant challenges in vehicle dynamics modeling, such as balancing model precision and computational efficiency at high speeds ( $>$ 280 km/h), where minor errors in modeling have severe consequences. Existing physics-based models for vehicle dynamics require elaborate testing setups and tuning, which are hard to implement, time-intensive, and cost-prohibitive. Conversely, purely data-driven approaches do not generalize well and cannot adequately ensure physical constraints on predictions. This letter introduces Deep Dynamics, a physics-constrained neural network (PCNN) for autonomous racecar vehicle dynamics modeling. It merges physics coefficient estimation and dynamical equations to accurately predict vehicle states at high speeds. A unique Physics Guard layer ensures internal coefficient estimates remain within their nominal physical ranges. Open-loop and closed-loop performance assessments, using a physics-based simulator and full-scale autonomous Indy racecar data, highlight Deep Dynamics as a promising approach for modeling racecar vehicle dynamics.
4 January 2021 in Natural Hazards
25 July 2022 in IEEE Robotics and Automation Letters
One of the main challenges in autonomous racing is to design algorithms for motion planning at high speed, and across complex racing courses. End-to-end trajectory synthesis has been previously proposed where the trajectory for the ego vehicle is computed based on camera images from the racecar. This is done in a supervised learning setting using behavioral cloning techniques. In this paper, we address the limitations of behavioral cloning methods for trajectory synthesis by introducing Differential Bayesian Filtering (DBF), which uses probabilistic Bézier curves as a basis for inferring optimal autonomous racing trajectories based on Bayesian inference. We introduce a trajectory sampling mechanism and combine it with a filtering process which is able to push the car to its physical driving limits. The performance of DBF is evaluated on the DeepRacing Formula One simulation environment and compared with several other trajectory synthesis approaches as well as human driving performance. DBF achieves the fastest lap time, and the fastest speed, by pushing the racecar closer to its limits of control while always remaining inside track bounds.
27 March 2026 in Journal of Intelligent & Robotic Systems
Accurate opponent trajectory prediction remains a significant challenge in autonomous racing, where vehicles often execute aggressive maneuvers and sudden acceleration changes. Transformer-based architectures like Motion Transformer (MTR) have shown promise for vehicle trajectory prediction in structured environments such as highways and urban roads. However, they, along with LSTM-enhanced methods like MixNet, face limitations in predicting trajectories under the extreme dynamics of autonomous racing, often leading to increased errors and computational inefficiencies. To address these challenges, we propose two novel methods. First, BézierMixNet extends MixNet by utilizing Bézier curves to more accurately represent trajectories through a combination of reference paths. Second, we introduce BARTé, which employs Composite Bézier Curves to achieve both computational efficiency and high prediction accuracy in dynamic racing scenarios. Our methods are evaluated on extensive datasets from the DeepRacing Formula One simulation and real-world data from the Indy Autonomous Challenge. Compared to state-of-the-art models, BARTé reduces average displacement error by approximately 4%, longitudinal error by 48% while reducing computation time by 90%, demonstrating significant advantages for high-speed autonomous racing.