Carlos Massera Filho

Longitudinal and lateral control for autonomous ground vehicles

Robust and stable control is a requirement for navigation of self-driving cars. Some approaches in the literature depend on a high number of parameters that are often difficult to estimate. A poor selection of these parameters often reduces considerably the efficiency of the control algorithms. In this paper we propose a simplified control system for autonomous vehicles that depends on a reduced number of parameters that can be easily set. This control system is composed of longitudinal and lateral controllers. The longitudinal controller is responsible for regulating the vehicles cruise velocity while the lateral controller steers the vehicles wheels for path tracking. Simulated and experimental tests have been carried out with the CaRINA II platform in the university campus with positive results.

Safe Optimization of Highway Traffic With Robust Model Predictive Control-Based Cooperative Adaptive Cruise Control

Road traffic crashes have been the leading cause of death among young people. Most of these accidents occur when the driver becomes distracted due to fatigue or external factors. Vehicle platooning systems, as cooperative adaptive cruise control, are one of the results of efforts devoted to the development of technologies for decreasing the number of road crashes and fatalities. Previous studies have suggested that such systems improve up to 273% highway traffic throughput and over 15% of fuel consumption if the clearance between vehicles in this class of roads can be reduced to 2 m. In this paper, we propose an approach that guarantees a minimum safety distance between vehicles taking into account the overall system delays and braking capacity of each vehicle. An

Simulation platform for cooperative vehicle systems

l\infty

Dynamic Inversion-Based Control for Front Wheel Drive Autonomous Ground Vehicles Near the Limits of Handling

-norm robust model predictive controller has been developed to guarantee the minimum safety distance is not violated due to uncertainties on the preceding vehicle behavior. A formulation for a lower bound clearance of vehicles inside a platoon is also proposed. Simulation results show the performance of the approach compared to a nominal controller when the system is subject to both modeled and unmodeled disturbances.

Driver Assistance Controller for Tire Saturation Avoidance up to the Limits of Handling

Road accidents is one of the major cause of deaths worldwide. Traffic is also one of the main problems in large cities. The development of intelligent transportation systems may contribute to the solution of both problems. Autonomous vehicles have been addressed by the academic community for decades and recently have been obtaining a considerable attention from the industry. One of the next steps for such technology is the development of communication systems that allow cooperative driving, which could improve the traffic efficiency. This paper proposes a simulation framework for the development of cooperative driving systems. Our framework is capable of accurate simulation of vehicles, sensors, complex environments, and wireless communication between vehicles. It is also compatible to well known robotic control middlewares, allowing portability of the develop code between real and simulated test vehicles. Results of cooperative adaptive cruise control with multiple vehicles are presented to validate the framework proposed.

Fast Metric Multi-Object Vehicle Tracking for Dynamical Environment Comprehension

Road traffic crashes are the leading cause of death among young people between 10 and 24 years old. Several safety systems for near the limits of handling scenarios such as Electronic Stability Control (ESC) and Active Front Steering (AFS) have been introduced in the past recent years. Vehicle stability in limit situations is extremely important to current driver assistance systems (DAS) and a safe control of Autonomous Ground Vehicles (AGV). This paper proposes an exponentially convergent dynamical model inversion based control for driving front wheel driven autonomous vehicles near tire-road friction saturation limits. Using a nonlinear planar bicycle model as basis, this control technique presented significative results under simulations even in cases of curvature non-continuities.

Lateral Control Based on Clothoids Curves for Heavy Category Ground Vehicles

Electronic Stability Control (ESC) reduced the number of fatal crashes in single vehicle accidents in 32%, a ESC unit corrects the vehicle yaw rate avoiding under-steering or over-steering in relation to the drives intent. In recent years Steer-by-Wire systems have been proposed for driver assistance systems, this allows the development of controllers capable of predicting and avoiding exceeding saturation limits. This paper proposes a Model Predictive Controller to safely handle longitudinal and lateral driver intents up to the limits of handling on front wheel driven vehicles.

Fast Metric Tracking by Detection System: Radar Blob and Camera Fusion

Tracking can be defined as the problem of estimating the trajectory of an object in image sequence as it moves around a scene. In other words, a tracker assigns consistent labels to the tracked objects in different frames of a video. One of the most widely used technique to this task is the Kalman filter. This paper presents a car tracker using Kalman filter and optical flow, which shows excellent results and low processing time.

A Simulation Framework for Multi-Vehicle Communication

This paper presents a lateral controller developed for autonomous heavy ground vehicles. This controller is responsible for regulating the direction of the steering wheel, and consequently, the frontal tires. It is also presented a brief introduction to works related to autonomous vehicles control, and an approach to obtain the kinematic model for this class of vehicles. The lateral control is designed based on clothoid curves and Fresnel integrals. Practical results are showed to illustrate the effectiveness of the approach proposed.

Pedestrian Classification Using K-means and Random Decision Forests

This article proposes a system that fuses radar and monocular vision sensor data in order to detect and classify on-road obstacles, like cars or not cars (other obstacles). The obstacle detection process and classification is divided into three stages, the first consist in reading radar signals and capturing the camera data, the second stage is the data fusion, and the third step is the classify the obstacles, aiming to differentiate the obstacles types identified by the radar and confirmed by the computer vision. In the detection task it is important to locate, measure, and rank the obstacles to be able to take adequate decisions and actions (e.g. Generate alerts, autonomously control the vehicle path), so the correct classification of the obstacle type and position, is very important, also avoiding false positives and/or false negatives in the classification task.