Browsing by Keyword "Task analysis"
Now showing 1 - 5 of 5
Results Per Page
Sort Options
Item Fast Real-Time Trajectory Planning Method with 3rd-Order Curve Optimization for Automated Vehicles(IEEE, 2020-09-20) Lattarulo, Ray; Perez, Joshue; CCAMAutomated driving (AD) is one of the fastest-growing tendencies in the Intelligent Transportation Systems (ITS) field with some interesting demonstrations and prototypes. Currently, the main research topics are aligned with vehicle communications, environment recognition, control, and decision-making. A real-time trajectory planning method for Automated vehicles (AVs) is presented in this paper; the contribution is part of AD’s decision-making module. This novel approach uses the properties of the 3er order Bézier curves to generate fast and reliable vehicle trajectories. Online execution and vehicle tracking capacities are considered on the approach. A feasible trajectory is selected based on the criteria: (i) the vehicle must be contained by a collision-free corridor given by an upper decision layer, (ii) the vehicle must be capable to track the generated trajectory, and (iii) the continuity of the path and curvature must be preserved in the joints. Our approach was tested considering a vehicle length (automated bus) of 12 meters. The scenario has the dimension of a real test location with multiple roundabouts.Item A Linear Model Predictive Planning Approach for Overtaking Manoeuvres Under Possible Collision Circumstances(IEEE, 2018-10-18) Lattarulo, Ray; He, Daniel; Perez, Joshue; Heß, Daniel; CCAMOvertaking is one of the most difficult tasks during driving. This manoeuvre demands good skills to accomplish it correctly. In the overtaking considering multiple vehicles (more than a couple) is necessary to understand, predict and coordinate future actions of the other participants. These reasons make it a significant scenario for testing in the connected and automated driving field, with the main goal of predicting safe future states. In this sense, this work presents an overtaking method based on a linear Model Predictive Control (MPC) approach, which considers multiple participants involved in the scenario. This method adapts dynamically the trajectory for the manoeuvre in case of unexpected situations. Some of these changes consider other vehicles coming on the opposite lane or variations on participants' driving decisions. Additionally, the system considers passengers' comfort, the vehicle physical constraints and lateral actions of the vehicle decoupled of the longitudinal ones to simplify the problem.Item Real-Time Trajectory Planning Method Based On N-Order Curve Optimization(Institute of Electrical and Electronics Engineers Inc., 2020-10-08) Lattarulo, Ray; Gonzalez, Leonardo; Perez, Joshue; Barbulescu, Lucian-Florentin; CCAMIn recent years, many functionalities were developed for Automated Vehicles (AVs) and some of them with close-to-market prototypes. A required topic is the generation of continuous trajectories that reduces the amount of discrete and pre-coded instructions while leading the vehicle safely. Consequently, this work presents a novel real-time trajectory planning approach based on numerical optimization of n-order Bézier curves and lane-based information. The generation of a feasible trajectory considers the vehicle dimension while driving into a lane-corridor. The nonlinear optimization problem was solved with the Bound Optimization BY Quadratic Approximation method (BOBYQA), and it uses the passengers' comfort, safety, and vehicle dynamics as constraints of the problem. The solution is validated in a simulation environment using a bus with a length of 12 meters. Moreover, the validation considered the roundabouts due to its complexity, nevertheless, the solution is scalable to other scenarios.Item A Review of Shared Control for Automated Vehicles: Theory and Applications(2020-12) Marcano, Mauricio; Diaz, Sergio; Perez, Joshue; Irigoyen, Eloy; CCAMThe last decade has shown an increasing interest on advanced driver assistance systems (ADAS) based on shared control, where automation is continuously supporting the driver at the control level with an adaptive authority. A first look at the literature offers two main research directions: 1) an ongoing effort to advance the theoretical comprehension of shared control, and 2) a diversity of automotive system applications with an increasing number of works in recent years. Yet, a global synthesis on these efforts is not available. To this end, this article covers the complete field of shared control in automated vehicles with an emphasis on these aspects: 1) concept, 2) categories, 3) algorithms, and 4) status of technology. Articles from the literature are classified in theory- and application-oriented contributions. From these, a clear distinction is found between coupled and uncoupled shared control. Also, model-based and model-free algorithms from these two categories are evaluated separately with a focus on systems using the steering wheel as the control interface. Model-based controllers tested by at least one real driver are tabulated to evaluate the performance of such systems. Results show that the inclusion of a driver model helps to reduce the conflicts at the steering. Also, variables such as driver state, driver effort, and safety indicators have a high impact on the calculation of the authority. Concerning the evaluation, driver-in-the-loop simulators are the most common platforms, with few works performed in real vehicles. Implementation in experimental vehicles is expected in the upcoming years.Item Transfer Learning and Online Learning for Traffic Forecasting under Different Data Availability Conditions: Alternatives and Pitfalls(Institute of Electrical and Electronics Engineers Inc., 2020-09-20) Manibardo, Eric L.; Laña, Ibai; Del Ser, Javier; IAThis work aims at unveiling the potential of Transfer Learning (TL) for developing a traffic flow forecasting model in scenarios of absent data. Knowledge transfer from high-quality predictive models becomes feasible under the TL paradigm, enabling the generation of new proper models with few data. In order to explore this capability, we identify three different levels of data absent scenarios, where TL techniques are applied among Deep Learning (DL) methods for traffic forecasting. Then, traditional batch learning is compared against TL based models using real traffic flow data, collected by deployed loops managed by the City Council of Madrid (Spain). In addition, we apply Online Learning (OL) techniques, where model receives an update after each prediction, in order to adapt to traffic flow trend changes and incrementally learn from new incoming traffic data. The obtained experimental results shed light on the advantages of transfer and online learning for traffic flow forecasting, and draw practical insights on their interplay with the amount of available training data at the location of interest.