Janito Vaqueiro Ferreira

A 2D/3D Vision Based Approach Applied to Road Detection in Urban Environments

This paper presents an approach for road detection based on image segmentation. This segmentation is resulted from merging 2D and 3D image processing data from a stereo vision system. The 2D layer returns a matrix containing pixels clusters based on the Watershed transform. Whereas the 3D layer return labels, that are classified by the V-Disparity technique, to free spaces, obstacles and non-classified area. Thus, a features descriptor for each cluster is composed with features from both layers. The road pattern recognition was performed by an artificial neural network, trained to obtain a final result from this features descriptor. The proposed work reports real experiments carried out in a challenging urban environment to illustrate the validity and application of this approach.

Advances on watershed processing on GPU architecture

This paper presents an overview on the advances of watershed processing algorithms executed on GPU architecture. The programming model, memory hierarchy and restrictions are discussed, and its influence on image processing algorithms detailed. The recently proposed algorithms of watershed transform for GPU computation are examined and briefly described. Its implementations are analyzed in depth and evaluations are made to compare them both on the GPU, against a CPU version and on two different GPU cards.

Real-time estimation of drivable image area based on monocular vision

Camera-based estimation of drivable image areas is still in evolution. These systems have been developed for improved safety and convenience, without the need to adapt itself to the environment. Machine Vision is an important tool to identify the region that includes the road in images. Road detection is the major task of autonomous vehicle guidance. In this way, this work proposes a drivable region detection algorithm that generates the region of interest from a dynamic threshold search method and from a drag process (DP). Applying the DP to estimation of drivable image areas has not been done yet, making the concept unique. Our system was has been evaluated from real data obtained by intelligent platforms and tested in different types of image texture, which include occlusion case, obstacle detection and reactive navigation.

Comprehensive performance analysis of road detection algorithms using the common urban Kitti-road benchmark

The navigation of an autonomous vehicle is a highly complex task and the dynamic environment is used as a source for reasoning. Road detection is a major issue in autonomous systems and advanced driving assistance systems applied for inner-city. Uncertainty may arise in environments with unmarked or weakly marked roads or poor lightning conditions. Moreover, when a common benchmark is not used, it is hard to decide which approach performs better on the road detection problem. This paper introduces a comprehensive performance analysis of two road recognition approaches using the urban Kitti-road benchmark. The first approach makes the extraction of a feature set based on statistical measures of 2D and 3D information from each superpixel. An Artificial Neural Network is used to detect the road pattern. The second approach extracts the feature set based on a multi-normalized histogram of Textons and Disptons for each superpixel. This feature set is used as a source for a Joint Boosting algorithm to model the road pattern. The proposed work presents a detailed evaluation highliting the pros and cons of each approach.

Evaluation of AUV fixed and vectorial propulsion systems with dynamic simulation and non-linear control

This paper presents an evaluation method for assessing different Autonomous Underwater Vehicle (AUV) propulsion systems (or topologies). This evaluation is based on solid/fluid dynamics simulation and uses a model-based control (feedback linearisation) to achieve the robotic mission assigned to the AUV. The objective of this work is to provide an evaluation relevant enough to be used for optimisation purposes. A model of an existing AUV (RSM) is proposed and detailed as well as two studied propulsion systems in competition: Fixed and vectorial thrusters. The main contribution of this work is to bring a consistent evaluation including models, tasks and a generic high-level control adapted to each system. Simulations of our AUV achieving a diving task are carried out for both propulsion systems. The simulated task analysis allows to evaluate the influence of the propulsive strategy over AUVs operability. Subsequently, numerical results are discussed in terms of energy efficiency and control.

Real-time Collision Risk Estimation based on Pearson's Correlation Coefficient

The perception of the environment is a major issue in autonomous robots. In our previous works, we have proposed a visual perception system based on an automatic image discarding method as a simple solution to improve the performance of a real-time navigation system. In this paper, we take place in the obstacle avoidance context for vehicles in dynamic and unknown environments, and we propose a new method for Collision Risk Estimation based on Pearsons Correlation Coefficient (PCC). Applying the PCC to real-time CRE has not been done yet, making the concept unique. This paper provides a novel way of calculating collision risk and applying it for object avoidance using the PCC. This real-time perception system has been evaluated from real data obtained by our intelligent vehicle.

Analysis of a Step-Based Watershed Algorithm Using CUDA

This paper proposes and develops a parallel algorithm for the watershed transform, with application on graphics hardware. The existing proposals are discussed and its aspects briefly analysed. The algorithm is proposed as a procedure of four steps, where each step performs a task using different approaches inspired by existing techniques. The algorithm is implemented using the CUDA libraries and its performance is measured on the GPU and compared to a sequential algorithm running on the CPU, achieving an average speed of twice the execution time of the sequential approach. This work improves on previous results of hybrid approaches and parallel algorithms with many steps of synchronisation and iterations between CPU and GPU.

Minimizing the edge buckling of the cold roll-forming process

In this work, the cold roll-forming process was numerically simulated by MSC SuperForm 2002 software based on finite element method. The strips were modelled with U-profile and a pre-punched hole located at the web zone was introduced. Two different configurations of the roll-forming mills were simulated, named as the Conventional and Curved. The conventional roll-forming mill was obtained by using the same diameters of the bottom rolls. The downhill roll-forming mill was achieved by increasing the diameters of the bottom rolls. This study investigated the occurrence of edge buckling in the hole lateral edge of the pre-punched sheet during the forming process. It could be concluded that, during the cold roll-forming process, reducing or even eliminating the compression stress in the web zone by the downhill roll-forming mill is possible to minimize occurrence of the edge buckling in the hole lateral edge of the pre-punched sheet during the forming process.

Experimental Vibration Characteristics of a Beam with Nonlinear Support Using Receptance Coupling Analysis

This papers presents the analytical and experimental results obtained of coupling a clamped beam with a nonlinear cubic stiffness joint. The analytical coupling results were obtained by using the Multi-Harmonic Nonlinear Receptance Coupling Approach (MUHANORCA) [4]. The experimental results were obtained by a sinusoidal excitation with a special force control algorithm where the level of the fundamental force is kept constant and the level of the harmonics is kept zero for all the frequencies measured.

Stereo Vision for Dynamic Urban Environment Perception Using Semantic Context in Evidential Grid

Uncertainty from urban environment arises not only by the imprecise pose estimation and noisy information in images but also by the lack of semantic information. This paper presents an approach to improve the perception capability of intelligent vehicles in complex urban environments. The method uses the meta-knowledge acquired from a built Semantic Context image and applies it on evidential grids constructed from the Stereo Vision. In the detection of semantic information problem, Texton and Dispton maps are used as a source to model a probabilistic joint boosting classifier. The evidential grids are based on occupancy grids, an approach based on the Dempster-Shafer theory that manages different sources of uncertainty arising in dynamic temporal scenes. An additional structure is created to handle a refined set of propositions originated from the meta-knowledge available. This structure bound with evidential conflict analysis enables an accurate detection of stationary and mobile objects in the perception grid. The proposed work reports real experiments carried out in a challenging urban environment using the KITTI benchmark in which meaningful evaluation can be done to illustrate the validity and application of this approach.