Browse Topic: Trajectory control
Autonomous vehicle navigation requires accurate prediction of driving path curvature to ensure smooth and safe trajectory planning. This paper presents a novel approach to curvature prediction using deep neural networks trained on GPS-derived ground truth data, rather than model predictions, providing a more accurate training signal that reflects actual vehicle motion. We develop a multi-modal neural network architecture with temporal GRU encoders that processes vision features, driver intent signals, historical curvature, and vehicle state parameters to predict curvature. A key innovation is the use of GPS-based actual curvature measurements computed from vehicle motion data (κ = ωz/v) as training supervision, enabling the model to learn from real-world driving patterns. The model is trained on 5,322 samples from real-world driving data collected on The University of Oklahoma’s Norman Campus using a Comma 3X device and a 2025 Nissan Leaf electric vehicle. Experimental results
In order to achieve fully autonomous driving, point to point autonomous navigation is the most important task. Most existing end-to-end models output a short-horizon path which makes the decision process hard to interpret and unreliable at intersections and complex driving scenarios. In this research, we build a navigation-integrated end-to-end path planner on top of an openpilot open source model. We created a navigation branch that encodes route polyline geometry, distance-to-next-maneuver, and high-level instructions and combines with path plan branch using residual blocks and feed-forward layers. By adding minimal parameters, new model keeps the original openpilot tasks unchanged and have the path output based on the navigation information. The model is trained on diverse urban scenes’ intersections, and it shows improved route performance in vehicle testing. The proposed model is validated in a Comma 3x device installed on a 2025 Nissan Leaf test vehicle. The road test results
This paper presents the design and implementation of a Semi-Autonomous Light Commercial Vehicle (LCV) capable of following a person while performing obstacle avoidance in urban and controlled environments. The LCV leverages its onboard 360-degree view camera, RTK-GNSS, Ultrasonic sensors, and algorithms to independently navigate the environment, avoiding obstacles and maintaining a safe distance from the person it is following. The path planning algorithm described here generates a secondary lateral path originating from the primary driving path to navigate around static obstacles. A Behavior Planner is utilized to decide when to generate the path and avoid obstacles. The primary objective is to ensure safe navigation in environments where static obstacles are prevalent. The LCV's path tracking is achieved using a combination of Pure Pursuit and Proportional-Integral (PI) controllers. The Pure Pursuit controller is utilized as lateral control to follow the generated path, ensuring
Semi-trailer trains are the main force of highway freight. In a complex environment with multiple vehicles, accidents are easily caused by complex structures and driver operation problems. Intelligent technology is urgently needed to improve safety. In view of the shortcomings of existing research on its dedicated models and algorithms, this paper studies the intelligent decision-making and trajectory planning of semi-trailer trains under multiple vehicles. A local trajectory planning method based on global path planning and Frenet coordinate decoupling based on the improved A* algorithm is proposed. The smooth weight transition function and B-spline curve are introduced to optimize the global path. The polynomial function is combined with the acceleration rate to optimize the local trajectory. TruckSim, Prescan and Simulink are used to build a joint simulation platform for multi-condition verification. The simulation results show that the search efficiency of the improved A* algorithm
Autonomous vehicle motion planning and control are vital components of next-generation intelligent transportation systems. Recent advances in both data- and physical model-driven methods have improved driving performance, yet current technologies still fall short of achieving human-level driving in complex, dynamic traffic scenarios. Key challenges include developing safe, efficient, and human-like motion planning strategies that can adapt to unpredictable environments. Data-driven approaches leverage deep neural networks to learn from extensive datasets, offering promising avenues for intelligent decision-making. However, these methods face issues such as covariate shift in imitation learning and difficulties in designing robust reward functions. In contrast, conventional physical model-driven techniques use rigorous mathematical formulations to generate optimal trajectories and handle dynamic constraints. Hybrid Data- and Physical Model-Driven Safe and Intelligent Motion Planning and
As the autonomy of ADAS features are moving from SAE level 0 autonomy to SAE level 5 autonomy of operation, reliance on AI/ML based algorithms in ADAS critical functions like perception, fusion and path planning are increasing predominantly. AI/ML based algorithms offer exceptional performance of the ADAS features, at the same time these advanced algorithms also bring in safety challenges as well. This paper explores the functional safety aspects of AI/ML based systems in ADAS functions like perception, object fusion and path planning, by discussing the safety requirements development for AI/ML systems, dataset safety life cycle, verification and validation of AI systems, and safety analysis used for AI systems. Among all the safety aspects listed above, emphasis is put on dataset safety lifecycle as that is not only the most important element for training ML based algorithms for ADAS usage, but also the most cumbersome and expensive. The safety characteristics associated with dataset
Items per page:
50
1 – 50 of 269