arXiv subjects

Denis F. Wolf

Publications and source records attributed to Denis F. Wolf.

10 recordsLinked to original sources

arXiv2023

Autonomous driving of trucks in off-road environment

Off-road driving operations can be a challenging environment for human conductors as they are subject to accidents, repetitive and tedious tasks, strong vibrations, which may affect their health in the long term. Therefore, they can benefit from a successful implementation of autonomous vehicle technology, improving safety, reducing labor costs and fuel consumption, and increasing operational efficiency. The main contribution of this paper is the experimental validation of a path tracking control strategy, composed of longitudinal and lateral controllers, on an off-road scenario with a fully-loaded heavy-duty truck. The longitudinal control strategy relies on a Non-Linear Model Predictive Controller (NMPC), which considers the path geometry and simplified vehicle dynamics to compute a smooth and comfortable input velocity, without violating the imposed constraints. The lateral controller is based on a Robust Linear Quadratic Regulator (RLQR), which considers a vehicle model subject to parametric uncertainties to minimize its lateral displacement and heading error, as well as ensure stability. Experiments were carried out using a fully-loaded vehicle on unpaved roads in an open-pit mine. The truck followed the reference path within the imposed constraints, showing robustness and driving smoothness.

eess.SY↗

arXiv2020

Tube-based Guaranteed Cost Model Predictive Control Applied to Autonomous Driving Up to the Limits of Handling

The development of control techniques to maintain vehicle stability under possible loss-of-control scenarios is essential to the safe deployment of autonomous ground vehicles in public scenarios. In this paper, we propose a tube-based guaranteed cost model predictive controller for autonomous vehicles able to avoid front and rear tire saturation and to track a provided reference trajectory up to the limits of handling of the vehicle. Such an approach ensures the vehicle will remain within its safe operational envelope; therefore, guaranteeing both stability and performance of the vehicle, including highly dynamic maneuvers that may be necessary for emergency conditions. We also propose a new conservative approximation of the nonlinear vehicle dynamics to a linear system subject to norm bounded multiplicative uncertainties and a new maximal robust controllable invariant set for vehicle dynamics. It consists of a larger feasible state-space region when compared to previously proposed invariant sets. Finally, we present both simulation and in-vehicle results of the performance of the proposed approach.

eess.SY↗

arXiv2020

Tube-based Guaranteed Cost Robust Model Predictive Control for Linear Systems Subject to Parametric Uncertainties

We propose a tube-based guaranteed cost model predictive controller considering a homothetic formulation for constrained linear systems subject to multiplicative structured norm-bounded uncertainties. It provides an upper bound to the general min-max model predictive control. The invariance property of the proposed tube holds for any arbitrary scaling. It yields a second-order cone programming problem which is less computationally expensive than standard semi-definite programming problems. We also present a numerical example with a comparative study among the proposed approach, an open-loop guaranteed cost model predictive controller, and a homothetic tube model predictive controller for linear difference inclusion systems.

eess.SY↗

arXiv2020

Image classification in frequency domain with 2SReLU: a second harmonics superposition activation function

Deep Convolutional Neural Networks are able to identify complex patterns and perform tasks with super-human capabilities. However, besides the exceptional results, they are not completely understood and it is still impractical to hand-engineer similar solutions. In this work, an image classification Convolutional Neural Network and its building blocks are described from a frequency domain perspective. Some network layers have established counterparts in the frequency domain like the convolutional and pooling layers. We propose the 2SReLU layer, a novel non-linear activation function that preserves high frequency components in deep networks. It is demonstrated that in the frequency domain it is possible to achieve competitive results without using the computationally costly convolution operation. A source code implementation in PyTorch is provided at: https://gitlab.com/thomio/2srelu

cs.CV↗

arXiv2019

Environment reconstruction on depth images using Generative Adversarial Networks

Robust perception systems are essential for autonomous vehicle safety. To navigate in a complex urban environment, it is necessary precise sensors with reliable data. The task of understanding the surroundings is hard by itself; for intelligent vehicles, it is even more critical due to the high speed in which the vehicle navigates. To successfully navigate in an urban environment, the perception system must quickly receive, process, and execute an action to guarantee both passenger and pedestrian safety. Stereo cameras collect environment information at many levels, e.g., depth, color, texture, shape, which guarantee ample knowledge about the surroundings. Even so, when compared to human, computational methods lack the ability to deal with missing information, i.e., occlusions. For many perception tasks, this lack of data can be a hindrance due to the environment incomplete information. In this paper, we address this problem and discuss recent methods to deal with occluded areas inference. We then introduce a loss function focused on disparity and environment depth data reconstruction, and a Generative Adversarial Network (GAN) architecture able to deal with occluded information inference. Our results present a coherent reconstruction on depth maps, estimating regions occluded by different obstacles. Our final contribution is a loss function focused on disparity data and a GAN able to extract depth features and estimate depth data by inpainting disparity images.

cs.CV↗

arXiv2019

Optimal Guaranteed Cost Control of Discrete-Time Linear Systems subject to Structured Uncertainties

In this paper, we propose state- and static output-feedback generalized guaranteed cost control (GCC) approaches for discrete-time linear systems subject to norm-bounded structured parametric uncertainties. This method enables the convex synthesis for a more general class of systems, where uncertainties are uncorrelated block diagonal, and no feed-through uncertainty is multiplicative with control input ones. It also provides necessary and sufficient conditions for state-feedback and sufficient conditions for static output-feedback. We also present a comparative study among the proposed controllers, standard Linear Quadratic Regulator, and Guaranteed Cost Controller found in the literature.

math.OC↗

arXiv2018

Driving Simulator Platform for Development and Evaluation of Safety and Emergency Systems

According to data from the United Nations, more than 3000 people have died each day in the world due to road traffic collision. Considering recent researches, the human error may be considered as the main responsible for these fatalities. Because of this, researchers seek alternatives to transfer the vehicle control from people to autonomous systems. However, providing this technological innovation for the people may demand complex challenges in the legal, economic and technological areas. Consequently, carmakers and researchers have divided the driving automation in safety and emergency systems that improve the driver perception on the road. This may reduce the human error. Therefore, the main contribution of this study is to propose a driving simulator platform to develop and evaluate safety and emergency systems, in the first design stage. This driving simulator platform has an advantage: a flexible software structure.This allows in the simulation one adaptation for development or evaluation of a system. The proposed driving simulator platform was tested in two applications: cooperative vehicle system development and the influence evaluation of a Driving Assistance System (\textit{DAS}) on a driver. In the cooperative vehicle system development, the results obtained show that the increment of the time delay in the communication among vehicles ($V2V$) is determinant for the system performance. On the other hand, in the influence evaluation of a \textit{DAS} in a driver, it was possible to conclude that the \textit{DAS'} model does not have the level of influence necessary in a driver to avoid an accident.

cs.CY↗

arXiv2017

Guaranteed Cost Approach to Robust Model Predictive Control of Uncertain Linear Systems

In this paper we propose a constrained guaranteed cost robust model predictive controller (GCMPC) for uncertain discrete time systems. This controller was developed based on a quadratic cost functional and guarantee robustness with respect to quadratically bound uncertainties. Such a class of problems is currently intractable by Min-Max Robust Model Predictive Controllers without polytopic approximations of the uncertainties. The proposed technique is computationally more efficient then an enumeration-based approach and requires only a Quadratically Constrained Quadratic Problem (QCQP) optimization, whereas LMI-based GCMPC approaches require a Semi-Definite Programming (SDP) optimization.

math.OC↗

arXiv2016

Guaranteed Cost Model Predictive Control-based Driver Assistance System for Vehicle Stabilization Under Tire Parameters Uncertainties

Road traffic crashes have been the leading cause of death among young people. Most of these accidents occur when the driver becomes distracted and a loss-of-control situation occurs. Steer-by-Wire systems were recently proposed as an alternative to mitigate such accidents. This technology enables the decoupling of the front wheel steering angles from the driver hand wheel angle and, consequently, the measurement of road/tire friction limits and the development of novel control systems capable of ensuring vehicle stabilization and safety. However, vehicle safety boundaries are highly dependent on tire characteristics which vary significantly with temperature, wear and the tire manufacturing process. Therefore, design of autonomous vehicle and driver assistance controllers cannot assume that these characteristics are constant or known. Thus, this paper proposes a Guaranteed Cost Model Predictive Controller Driver Assistance System able to avoid front and rear tire saturation and to track the drivers intent up to the limits of handling for a vehicle with uncertain tire parameters. Simulation results show the performance of the proposed approach under time-varying uniformly distributed disturbances.

math.OC↗

arXiv2016

Safely Optimizing 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 such as Cooperative Adaptive Cruise Control (CACC) are one of the results of the effort devoted to the development of technologies for decreasing the number of road crashes and fatalities. Previous studies have suggested such systems improve up to 273\% highway traffic throughput and fuel consumption in more than 15\% if the clearance between vehicles in this class of roads can be reduced to 2 meters. This paper proposes an approach that guarantees a minimum safety distance between vehicles taking into account the overall system delays and braking capacity of each vehicle. A $l\infty$-norm Robust Model Predictive Controller (RMPC) is developed to guarantee the minimum safety distance is not violated due to uncertainties on the lead vehicle behavior. A formulation for a lower bound clearance of vehicles inside a platoon is also proposed. Simulation results show the performance of the proposed approach compared to a nominal controller when the system is subject to both modeled and unmodeled disturbances.

math.OC↗