In the typical autonomous driving stack, planning and control systems represent two of the most crucial components in which data retrieved by sensors and processed by perception algorithms are used to implement a safe and comfortable self-driving behavior. In particular, the planning module predicts the path the autonomous car should follow taking the correct high-level maneuver, while control systems perform a sequence of low-level actions, controlling steering angle, throttle and brake. In this work, we propose a model-free Deep Reinforcement Learning Planner training a neural network that predicts both acceleration and steering angle, thus obtaining a single module able to drive the vehicle using the data processed by localization and perception algorithms on board of the self-driving car. In particular, the system that was fully trained in simulation is able to drive smoothly and safely in obstacle-free environments both in simulation and in a real-world urban area of the city of Parma, proving that the system features good generalization capabilities also driving in those parts outside the training scenarios. Moreover, in order to deploy the system on board of the real self-driving car and to reduce the gap between simulated and real-world performances, we also develop a module represented by a tiny neural network able to reproduce the real vehicle dynamic behavior during the training in simulation.

Navigating through intersections is one of the main challenging tasks for an autonomous vehicle. However, for the majority of intersections regulated by traffic lights, the problem could be solved by a simple rule-based method in which the autonomous vehicle behavior is closely related to the traffic light states. In this work, we focus on the implementation of a system able to navigate through intersections where only traffic signs are provided. We propose a multi-agent system using a continuous, model-free Deep Reinforcement Learning algorithm used to train a neural network for predicting both the acceleration and the steering angle at each time step. We demonstrate that agents learn both the basic rules needed to handle intersections by understanding the priorities of other learners inside the environment, and to drive safely along their paths. Moreover, a comparison between our system and a rule-based method proves that our model achieves better results especially with dense traffic conditions. Finally, we test our system on real world scenarios using real recorded traffic data, proving that our module is able to generalize both to unseen environments and to different traffic conditions.

Deep Reinforcement Learning has proved to be able to solve many control tasks in different fields, but the behavior of these systems is not always as expected when deployed in real-world scenarios. This is mainly due to the lack of domain adaptation between simulated and real-world data together with the absence of distinction between train and test datasets. In this work, we investigate these problems in the autonomous driving field, especially for a maneuver planning module for roundabout insertions. In particular, we present a system based on multiple environments in which agents are trained simultaneously, evaluating the behavior of the model in different scenarios. Finally, we analyze techniques aimed at reducing the gap between simulated and real-world data showing that this increased the generalization capabilities of the system both on unseen and real-world scenarios.

Video sensors based on CMOS technology represent the vast majority of the adopted solutions for image acquisition systems equipped in unmanned vehicles and Advanced Driver Assistance Systems. These sensors typically employ a rolling shutter readout scheme, which causes image distortion in case of motion. This paper investigates how rolling shutter effects influence the performance of Semi-Global Matching, a state of the art dense stereo matching algorithm. In order to exhaustively evaluate those effects, the algorithm is computed over sequences acquired with a dedicated image acquisition system composed by two calibrated and synchronized stereo cameras, the first provided with global shutter sensors and the second with rolling shutter sensors. A pixel-wise comparison between the computed Disparity Space Images allows to determine the error in a field automotive test. An effective correction method is then applied to the rolling shutter Disparity Space Image, aiming to reduce the error and therefore to improve accurate position estimation of obstacles in automotive scenarios.

Obstacles detection is used nowdays for a number of road safety applications, increasing the drivers awareness in potential dangerous situations. A reliable and robust obstacles detection continues to be largely investigated and still remains an open challenge, especially for difficult scenarios and in general cases, with loosened constraints and multiple simultaneous use-cases. This work presents an obstacles detection, tracking and fusion algorithm which allows to reconstruct the environment surrounding the vehicle. While the techniques used for the detection are well-known in literature, the improvements introduced by this paper regard the data association and tracking approach of heterogeneous sensors observations. An innovative multi-dimensional structure based on association costs originating from a classifier provides an optimal solution to the association problem with respect to the total association cost. An Unscented Kalman Filter (UKF) managing a variable number of observations, arbitrarily composable, allows to correctly address the combined tracking and fusion challenge. The results, obtained on a public benchmark, show improvements with respect to state of the art systems.

Three-dimensional environment perception is one of the most important tasks for an autonomous vehicle. Map-based approaches play a fundamental role in the representation of vehicle surroundings, allowing several perception features, such as obstacle detection or road classification. However, benchmarks available in literature do not allow to evaluate the accuracy of these discrete representations, focusing only on the results downstream the maps. The proposed system uses a stochastic approach to evaluate a generic discrete representation of a three-dimensional world. The evaluation process consists in comparing a local perceived representation with the corresponding previously computed ground truth. The ground truth is automatically generated exploiting either accurate depth sensing and precise localization information. A test case is proposed, using stereo vision data and Digital Elevation Maps.

This paper presents the design, the setup, and the architecture of a new class of autonomous vehicles prototype. The car is equipped with 26 cameras, divided into 13 stereo pairs. Four stereo systems are dedicated to the reconstruction of the near area surrounding the vehicle, other nine are dedicated to the 3D reconstruction of the far driving area. Additionally four laserscanners, and a high performance GPS/IMU unit provide more information on the ground truth to measure the validity of the data obtained with vision. The autonomous driving system, the perception system, and the processing system have been installed in the vehicle taking extra care for an as-clean-as-possible integration with te aim of setting a new state of the art level for this kind of setup. A Human Machine Interface, allows the driver to control all the functions of the systems using a touch screen. A high precision synchronization system coupled with a custom software architecture allows to obtain recordings from all sensors installed.

This paper presents an autonomous driving test held in Parma on urban roads and freeways open to regular traffic. During this test, the vehicle not only performed simple maneuvers, but it had to cope with complex driving scenarios as well, including roundabouts, junctions, pedestrian crossings, freeway junctions, and traffic lights. The test demonstrated the ability of the current technology to manage real situations and not only the well-structured and predictable ones. A comparison of milestones, challenges, and key results in autonomous driving is presented to highlight the novelty and the specific purpose of the test. The whole system is described: the vehicle; the software architecture; details about high-, medium-, and low-level control; and details about perception algorithms. A conclusion highlights the achieved results and draws possible directions for future development.

This paper presents a monocular algorithm for front and rear vehicle detection, developed as part of the FP7 V-Charge project's perception system. The system is made of an AdaBoost classifier with Haar Features Decision Stump. It processes several virtual perspective images, obtained by un-warping 4 monocular fish-eye cameras mounted all-around an autonomous electric car. The target scenario is the automated valet parking, but the presented technique fits well in any general urban and highway environment. A great attention has been given to optimize the computational performance. The accuracy in the detection and a low computation costs are provided by combining a multiscale detection scheme with a Soft-Cascade classifier design. The algorithm runs in real time on the project's hardware platform. The system has been tested on a validation set, compared with several AdaBoost schemes, and the corresponding results and statistics are also reported.

The presence of autonomous vehicles on public roads is becoming a reality. In the last 10 years, autonomous prototypes have been confined in controlled or isolated environments, but new traffic regulations for testing and direct automotive companies interests are moving autonomous vehicles tests on real roads. This paper presents a test on public urban roads and freeways that was held in Parma on July 12, 2013. This was the first test in open public urban roads with nobody behind the steering wheel: the vehicle had to cope with roundabouts, junctions, pedestrian crossings, freeway junctions, traffic lights, and regular traffic. The vehicle setup, the software architecture, and the route are here presented together with some results and possible future improvements.

Path estimation is a big challenge for autonomous vehicle navigation, especially in unknown, dynamic environments, when road characteristics change often. 3D terrain information (e.g. stereo cameras) can provide useful hints about the traversability cost of certain regions. However, when the terrain tends to be flat and uniform, it is difficult to identify a better path using 3D map solely. In this scenario the use of a priori knowledge on the expected road's visual characteristics can support detection, but it has the drawback of being not robust to environmental changes. This paper presents a path detection method that mixes together 3D mapping and visual classification, trying to learn, in real time, the actual road characteristics. An on-line learning of visual characteristics is implemented to feedback a terrain classifier, so that the road characteristics are updated as the vehicle moves. The feedback data are taken from a 3D traversability cost map, which provides some hints on traversable and non-traversable regions. After several re-training cycles the algorithm converges on a better separation of the path and non-path regions. The fusion of both 3D traversability cost and visual characteristics of the terrain yields a better estimation when compared with either of these methods solely.

Autonomous Ground Vehicles designed for extreme environments (e.g mining, constructions, defense, exploration applications) require a reliable estimation of terrain traversability, in terms of both terrain slope and obstacles presence. In this paper we present a new technique to build, in real time and only from a 3D points cloud, a dense terrain elevation map able to: 1) provide slope estimation; 2) provide a reference for segmenting points into terrain's inliers and outliers, to be then used for obstacles detection. The points cloud is first smartly sampled into a 2.5 grid map, then samples are fitted into a rational B-Spline surface by means of re-weighted least square fitting and equalization. To meet an extensive range of extreme off-road scenarios, no assumptions on vehicle pose are made and no road infrastructure or a-priori knowledge about terrain appearance and shape is required. The algorithm runs in real time; it has been tested on one of VisLab's AGVs using a modified SGM-based stereo system as 3D data source.

This paper presents a new vehicle guidance paradigm based on high level commands together with a new system architecture that can handle different levels of automation. First, an overview of different automation levels spanning from manual driving to fully autonomous driving is given, then a detailed description of the digital driving paradigm is also shown. This includes a detailed description of the maneuver level planner which is based on a new concept of planning based on checkpoints. Then, an overview of the new architecture is given, describing is key aspects. An analysis of the proposed paradigm, with the description of a preliminary tests in real scenarios concludes this paper.

Research on pedestrian detection system still presents a lot of space for improvements, both on speed and detection accuracy. This paper presents a full implementation of a pedestrian detection system, using a part-based classification for the candidates identification and a feature based tracking for increasing the result robustness. The novelty of this approach relies on the use of part-based approach with a combination of Haar-cascade and HOG-SVM. Tests have been conducted using standard datasets showing results aligned with those of the other state-of-the-art systems available in literature. Real world tests also show high speed performance.

This paper presents the vision of the Artificial Vision and Intelligent Systems Laboratory (VisLab) on future automated vehicles, ranging from sensor selection up to their extensive testing. VisLab's design choices are explained using the BRAiVE autonomous vehicle prototype as an example. BRAiVE, which is specifically designed to develop, test, and demonstrate advanced safety applications with different automation levels, features a high integration level and a low-cost sensor suite, which are mainly based on vision, as opposed to many other autonomous vehicle implementations based on expensive and invasive sensors. The importance of performing extensive tests to validate the design choices is considered to be a hard requirement, and different tests have been organized, including an intercontinental trip from Italy to China. This paper also presents the test, the main challenges, and the vehicles that have been specifically developed for this test, which was performed by four autonomous vehicles based on BRAiVE's architecture. This paper also includes final remarks on VisLab's perspective on future vehicles' sensor suite.

Detecting pedestrians is still a challenging task for automotive vision systems due to the extreme variability of targets, lighting conditions, occlusion, and high-speed vehicle motion. Much research has been focused on this problem in the last ten years and detectors based on classifiers have gained a special place among the different approaches presented. This paper presents a state-of-the-art pedestrian detection system based on a two-stage classifier. Candidates are extracted with a Haar cascade classifier trained with the Daimler Detection Benchmark data set and then validated through a part-based histogram-of-oriented-gradient (HOG) classifier with the aim of lowering the number of false positives. The surviving candidates are then filtered with feature-based tracking to enhance the recognition robustness and improve the results' stability. The system has been implemented on a prototype vehicle and offers high performance in terms of several metrics, such as detection rate, false positives per hour, and frame rate. The novelty of this system relies on the combination of a HOG part-based approach, tracking based on a specific optimized feature, and porting on a real prototype.

Trustworthiness is a key point when dealing with vehicle safety applications. In this paper an approach to a real application is presented, able to fulfill the requirements of such demanding applications. Most of commercial sensors available nowadays are usually designed to detect front vehicles but lack the ability to detect overtaking vehicles. The work presented here combines the information provided by two sensors, a Stop&Go radar and a camera. Fusion is done by using the unprocessed information from the radar, and computer vision based on optical flow. The basic capabilities of the commercial systems are upgraded giving the possibility to improve the front vehicles detection system, by detecting overtaking vehicles with a high positive rate.

This paper presents an environment-detection-and-mapping algorithm for autonomous driving that is provided in real time and for both rural and off-road environments. Environment-detection-and-mapping algorithms have been designed to consist of two parts: 1) lane, pedestrian-crossing, and speed-bump detection algorithms using cameras and 2) obstacle detection algorithm using LIDARs. The lane detection algorithm returns lane positions using one camera and the vision module “VisLab Embedded Lane Detector (VELD),” and the pedestrian-crossing and speed-bump detection algorithms return the position of pedestrian crossings and speed bumps. The obstacle detection algorithm organizes data from LIDARs and generates a local obstacle position map. The designed algorithms have been implemented on a passenger car using six LIDARs, three cameras, and real-time devices, including personal computers (PCs). Vehicle tests have been conducted, and test results have shown that the vehicle can reach the desired goal with the proposed algorithm.

Obstacle detection by means of stereo-vision is a fundamental task in computer vision, which has spurred a lot of research over the years, especially in the field of vehicular robotics. The information provided by this class of algorithms is used both in driving assistance systems and in autonomous vehicles, so the quality of the results and the processing times become critical, as detection failures or delays can have serious consequences. The obstacle detection system presented in this paper has been extensively tested during VIAC, the VisLab Intercontinental Autonomous Challenge [1], [2], which has offered a unique chance to face a number of different scenarios along the roads of two continents, in a variety of conditions; data collected during the expedition has also become a reference benchmark for further algorithm improvements.

This paper presents the preliminary results of VIAC, the VisLab Intercontinental Autonomous Challenge, a test of autonomous driving along an unknown route from Italy to China. It took 3 months to run the entire test; all data have been logged, including all data generated by the sensors, vehicle data, and GPS info. This huge amount of information has been packed during the trip, compressed, and transferred back to Parma for further processing. This data is now ready for a deep analysis of the various systems performance, with the aim of virtually running the whole trip multiple times with improved versions of the software. This paper discusses some preliminary figures obtained by the analysis of the data collected during the test. More information will be generated by a deeper analysis, which will take additional time, being the data about 40 terabyte in size.

This paper presents the design issues that were considered for the equipment of 4 identical autonomous vehicles that will drive themselves without human intervention on an intercontinental route for more than 13,000 km. Autonomous vehicles have been demonstrated able to reach the end of a 220 miles off-road trail (in the DARPA Grand Challenge), to negotiate traffic and obey traffic rules (in the DARPA Urban Challenge), but no one ever tested their capabilities on a long, intercontinental trip and stressed their systems for 3 months in a row. This paper presents the technological challenge of a set of vehicles that will run the VisLab Intercontinental Autonomous Challenge (VIAC). The challenge is scheduled to take place during the 2010 World Expo in Shanghai, China (and precisely from July 10, 2010 to Oct 10, 2010). Being currently under preparation, this paper focuses on the development, the vehicles' technical details, and the challenge itself. Other following papers will describe the outcome of the challenge and its results.

Reliable perception of terrain slope and terrain traversability is a key-feature for any off-road unmanned ground vehicle, as well as for any Driver Assistance Systems designed to work in extreme environments, like mining. In this paper we want to present an innovative technique to build a 3D elevation map of the traversable terrain from a world's 3D dense data set, in real time. The 3D points are grouped into lateral and longitudinal equally spaced slices, then they are projected onto the corresponding slices' reference planes. The projections are then analyzed by a biologically inspired Optimization Algorithm able to segment points into terrains inlier and outlier; the resulting 2D terrain slopes represent an optimal terrain approximation along each slice. Finally, the 2D approximations are merged together, to create the overall 3D terrain surface. The algorithm has been successfully tested with 3D data provided by a stereo camera system mounted on a Cat®wheel loader operating in a mining environment.

This paper presents the TerraMax vision systems used during the 2007 DARPA Urban Challenge. First, a description of the different vision systems is provided, focusing on their hardware configuration, calibration method, and tasks. Then, each component is described in detail, focusing on the algorithms and sensor fusion opportunities: obstacle detection, road marking detection, and vehicle detection. The conclusions summarize the lesson learned from the developing of the passive sensing suite and its successful fielding in the Urban Challenge.

This paper presents an application of a pedestrian-detection system aimed at localizing potentially dangerous situations under specific urban scenarios. The approach used in this paper differs from those implemented in traditional pedestrian-detection systems, which are designed to localize all pedestrians in the area in front of the vehicle. Conversely, this approach searches for pedestrians in critical areas only. The environment is reconstructed with a standard laser scanner, whereas the following check for the presence of pedestrians is performed due to the fusion with a vision system. The great advantages of such an approach are that pedestrian recognition is performed on limited image areas, therefore boosting its timewise performance, and no assessment on the danger level is finally required before providing the result to either the driver or an onboard computer for automatic maneuvers. A further advantage is the drastic reduction of false alarms, making this system robust enough to control nonreversible safety systems.

The paper discusses the journal efforts on intelligent transport systems. 2009 marks the end of the first decade of our IEEE Transactions on Intelligent Transportation Systems (T-ITS). Over the past ten years, we have experienced various ups and downs, but we survived, and now, we thrive. In this last issue of the passing decade, the Founding, Past, and Current Editors in Chief would like to join everyone in celebrating the tenth anniversary of this great journal. We would like to take this opportunity to share some ideas and reflections on the past, present, and future of this publication.

This paper describes a vehicle detection system fusing radar and vision data. Radar data are used to locate areas of interest on images. Vehicle search in these areas is mainly based on vertical symmetry. All the vehicles found in different image areas are mixed together, and a series of filters is applied in order to delete false detections. In order to speed up and improve system performance, guard rail detection and a method to manage overlapping areas are also included. Both methods are explained and justified in this paper. The current algorithm analyzes images on a frame-by-frame basis without any temporal correlation. Two different statistics, namely: 1) frame based and 2) event based, are computed to evaluate vehicle detection efficiency, while guard rail detection efficiency is computed in terms of time savings and correct detection rates. Results and problems are discussed, and directions for future enhancements are provided

This paper presents a sensor fusion model developed for the 2005 Grand Challenge competition, an autonomous ground vehicle race across the Mojave desert organized by DARPA. The two sensors used in this work are a stereo vision camera pair and an ALASCA laserscanner. An algorithm to filter laserscanner's raw scan data and to remove ground reflections is also presented. Several tests were made to prove the reliability of this method, that has proved to be useful to extract the information required by the race. Fusion was performed both at a low and medium level: terrain slope, detected with stereo vision, was used to correct pitch information of laserscanner raw data. Object segmentation is applied on a bird view bitmap where each pixel represents a square area of the world in front of the vehicle; this bitmap is obtained from the fusion of the ones generated by each sensor

This paper presents a stereo vision system for the detection and distance computation of a preceding vehicle. It is divided in two major steps. Initially, a stereo vision-based algorithm is used to extract relevant three-dimensional (3-D) features in the scene, these features are investigated further in order to select the ones that belong to vertical objects only and not to the road or background. These 3-D vertical features are then used as a starting point for preceding vehicle detection; by using a symmetry operator, a match against a simplified model of a rear vehicle's shape is performed using a monocular vision-based approach that allows the identification of a preceding vehicle. In addition, using the 3-D information previously extracted, an accurate distance computation is performed.

The last few decades have witnessed the birth and growth of a new sensibility to transportation efficiency. In particular the need for efficient and improved people and goods mobility has pushed researchers to address the problem of intelligent transportation systems. This paper surveys the most advanced approaches to (partial) customization of the road following task, using on-board systems based on artificial vision. The functionalities of lane detection, obstacle detection and pedestrian detection are described and classified, and their possible application in future road vehicles is discussed.

Describes the image processing techniques designed to localize the tracks of snowcats for the automation of transportation of goods and people during the Italian scientific missions in Antarctica. The final goal is to enable a snowcat to automatically follow the preceding one in a train-like fashion. A camera is used to acquire images of the scene; the image sequence is analyzed by a computer vision system which identifies the tracks and produces a high level description of the scene. This result is then forwarded to a further software module in charge of the control of the snowcat movement. A further optional representation, in which markers highlighting the tracks are superimposed onto the acquired image, is transmitted to a human supervisor located off board. This system has been tested in the Italian test site and was under testing in the South Pole during the early 2002 Italian scientific mission. The paper also briefly describes an alternative solution based on an evolutionary approach.

This paper presents a new motion planning primitive to be used for the iterative steering of vision-based autonomous vehicles. This primitive is a parameterized quintic spline, denoted as /spl eta/-spline, that allows interpolating an arbitrary sequence of points with overall second-order geometric (G/sup 2/-) continuity. Issues such as completeness, minimality, regularity, symmetry, and flexibility of these G/sup 2/-splines are addressed in the exposition. The development of the new primitive is tightly connected to the inversion control of nonholonomic car-like vehicles. The paper also exposes a supervisory strategy for iterative steering that integrates feedback vision data processing with the feedforward inversion control.

In this paper a calibration method for on-board cameras used on the ARGO autonomous vehicle is presented. A number of markers have been placed on the vehicle's hood, which is framed by the vision system. Due to the knowledge of the markers' position it is possible to compute the camera position and orientation with respect to the vehicle. By using fast computations, this procedure, of basic importance when the camera head has pan-tilt capabilities, can be performed during autonomous driving, without slowing down normal operations.

Presents the methods for sensing obstacles and vehicles implemented on the University of Parma experimental vehicle (ARGO). The ARGO project is briefly described along with its main objectives; the prototype vehicle and its functionalities are presented. The perception of the environment is performed through the processing of images acquired from the vehicle. Details about the stereo vision-based detection of generic obstacles are given, along with a measurement of the performance of the method; then a new approach for leading vehicles detection is described, relying on symmetry detection in monocular images. The paper concludes with a description of the current implementation of the control system, based on a gain scheduled controller, which allows the vehicle to follow the road or other vehicles.

A number of different algorithms have been described in the literature for the decomposition of both convex binary morphological structuring elements and a specific subset of nonconvex ones. Nevertheless, up to now no deterministic solutions have been found to the problem of decomposing arbitrarily shaped structuring elements. This work presents a new stochastic approach based on genetic algorithms, in which no constraints are imposed on the shape of the initial structuring element nor assumptions are made on the elementary factors, which are selected within a given set.

This paper describes the generic obstacle and lane detection system (GOLD), a stereo vision-based hardware and software architecture to be used on moving vehicles to increment road safety. Based on a full-custom massively parallel hardware, it allows to detect both generic obstacles (without constraints on symmetry or shape) and the lane position in a structured environment (with painted lane markings) at a rate of 10 Hz. Thanks to a geometrical transform supported by a specific hardware module, the perspective effect is removed from both left and right stereo images; the left is used to detect lane markings with a series of morphological filters, while both remapped stereo images are used for the detection of free-space in front of the vehicle. The output of the processing is displayed on both an on-board monitor and a control-panel to give visual feedbacks to the driver. The system was tested on the mobile laboratory (MOB-LAB) experimental land vehicle, which was driven for more than 3000 km along extra-urban roads and freeways at speeds up to 80 km/h, and demonstrated its robustness with respect to shadows and changing illumination conditions, different road textures, and vehicle movement.

This work presents a system for obstacle detection in a pair of images acquired by a stereo vision device installed on a moving vehicle. The whole system is structured in a pipeline of two different computational engines: a massively parallel architecture, PAPRICA, devoted to low-level image processing and a traditional serial architecture running medium-level tasks. A geometrical transformation, based on the assumption of a flat road in front of the vehicle, is performed to remove the perspective effect from both images. The difference between the results is used for the detection of free-space in front of the vehicle, thus allowing to avoid the high computational tasks involved in traditional stereo vision approaches; the geometrical transformation is performed by a specific hardware device integrated in PAPRICA architecture. The system was tested on the MOB-LAB experimental land vehicle, which was driven for more than 3000 km along extra-urban roads and freeways at speeds up to 80 km/h, and demonstrated its robustness with respect to shadows and changing illumination conditions, different road textures, and vehicle movement.

This paper presents an image reorganization procedure aimed to speed-up and simplify the detection of road boundaries in automotive road following applications. Some results of road markings detection in images acquired from the MOB-LAB land vehicle are presented, together with the description of the current and future implementations of the reorganization procedure on the PAPRICA system. The data reorganization process offers the possibility of designing and implementing on a SIMD processor an extremely simple and robust low-level algorithm for the detection of road markings.

The main aim of this work is the development of a vision-based road detection system fast enough to cope with the difficult real-time constraints imposed by moving vehicle applications. The hardware platform, a special-purpose massively parallel system, has been chosen to minimize system production and operational costs. This paper presents a novel approach to expectation-driven low-level image segmentation, which can be mapped naturally onto mesh-connected massively parallel SIMD architectures capable of handling hierarchical data structures. The input image is assumed to contain a distorted version of a given template; a multiresolution stretching process is used to reshape the original template in accordance with the acquired image content, minimizing a potential function. The distorted template is the process output.

Presents a system for the extraction of road boundaries from an image taken in an out-of-town environment. In this application, computational speed and performance play a fundamental role in the selection of the hardware platform and the design of algorithms. The algorithm has been designed to be implemented on a special-purpose mesh-connected SIR ID architecture, PAPRICA, which will be fitted to the vehicle. This presentation focuses on the algorithms and in particular on processing speed.

The paper describes the architecture of an image coprocessor based on a linear array of elementary processors. The system has been designed for real-time vision applications and embeds a direct I/O interface to a camera device. The instruction set of the elementary processors is based on morphological match operators on a reduced 5/spl times/5 neighborhood. Two additional communication mechanisms, one fixed and one programmable, provide both global and multiple communication channels among processing elements which are not directly connected.>

This work presents a Computer Vision system for road boundary detection in automotive applications. Images are processed by a multiresolution algorithm, driven by a-priori knowledge through a top-down control. In order to face the hard real-time constraints of automotive tasks, a special purpose massively parallel computer architecture, PAPRICA, has been developed. The whole system is currently operative on MOB-LAB mobile laboratory: a land vehicle integrating the results of the activities of the Italian PROMETHEUS units. The basis of the algorithm is discussed using the formal tools of mathematical morphology, while the choice of the computing architecture and of the computational paradigm is explained. The generality of the presented approach allows its use also to solve similar problems, namely to detect features exploiting a long-distance correlation, such as the road boundaries in vehicular applications.>

The PAPRICA project started in 1988 as an experimental VLSI architecture devoted to the efficient computation of data with two-dimensional structure. The main goal of the project is to develop a subsystem that could operate as an attached processing unit to a standard workstation and in perspective as a specialized processing module in dedicated systems devoted to low level image analysis, cellular neural networks emulation, DRC algorithms. The architecture has been extensively used for basic low level image analysis tasks up to optical flow computation and feature tracking, showing encouraging performance even in the first prototype version. The authors discuss the actual implementation and present a critical analysis of the project, allowing to identify some crucial points of PAPRICA design (and of array processors in general) that must be carefully considered in the case of redesign.>