Rafael Toledo-Moreo

High-Integrity IMM-EKF-Based Road Vehicle Navigation With Low-Cost GPS/SBAS/INS

User requirements for the performance of Global Navigation Satellite System (GNSS)-based road applications have been significantly increasing in recent years. Safety systems based on vehicle localization, electronic fee-collection systems, and traveler information services are just a few examples of interesting applications requiring onboard equipment (OBE) capable of offering a high available accurate position, even in unfriendly environments with low satellite visibility such as built-up areas or tunnels and at low cost. In addition to that, users and service providers demand from the OBEs not only accurate continuous positioning but integrity information of the reliability of this position as well. Specifically, in life-critical applications, high-integrity monitored positioning is absolutely required. This paper presents a solution based on the fusion of GNSS and inertial sensors (a Global Positioning System/satellite-based augmentation system/inertial navigation system integrated system) running an extended Kalman filter combined with an interactive multimodel method (IMM-EKF). The solution developed in this paper supplies continuous positioning in marketable conditions and a meaningful trust level of the given solution. A set of tests performed in controlled and real scenarios proves the suitability of the proposed IMM-EKF implementation as compared with low-cost GNSS-based solutions, dead reckoning systems, single-model EKF, and other filtering approaches of the current literature.

Lane-Level Integrity Provision for Navigation and Map Matching With GNSS, Dead Reckoning, and Enhanced Maps

Lane-level positioning and map matching are some of the biggest challenges for navigation systems. Additionally, in safety applications or in those with critical performance requirements (such as satellite-based electronic fee collection), integrity becomes a key word for the navigation community. In this scenario, it is clear that a navigation system that can operate at the lane level while providing integrity parameters that are capable of monitoring the quality of the solution can bring important benefits to these applications. This paper presents a pioneering novel solution to the problem of combined positioning and map matching with integrity provision at the lane level. The system under consideration hybridizes measurements from a global navigation satellite system (GNSS) receiver, an odometer, and a gyroscope, along with the road information stored in enhanced digital maps, by means of a multiple-hypothesis particle-filter-based algorithm. A set of experiments in real environments in France and Germany shows the very good results obtained in terms of positioning, map matching, and integrity consistency, proving the feasibility of our proposal.

A Cooperative Approach to Traffic Congestion Detection With Complex Event Processing and VANET

Currently, distributed traffic information systems have come up as one of the most important approaches for detecting traffic flow problems on a road. For that purpose, they usually make use of the location information that vehicles share among them through periodical messages that are transmitted across a vehicular ad hoc network (VANET). This paper puts forward an event-driven architecture (EDA) as a novel mechanism to get insight into VANET messages to detect different levels of traffic jams; furthermore, it also takes into account environmental data that come from external data sources, such as weather conditions. The proposed EDA has been developed through the complex-event-processing technology. Simulation tests show that the proposed mechanism can detect traffic congestions, which involve different numbers of lanes and lengths with short delay.

Creating Enhanced Maps for Lane-Level Vehicle Navigation

The concept of enhanced maps (Emaps) was introduced with one main objective: It should characterize roads, first, with more completeness and, second, with more accuracy than standard maps to fulfill the requirements of new challenging road safety applications and advanced driver-assistance systems (ADAS). This paper introduces a paradigm for Emap definition and creation on which every road lane is represented and topologically connected to the rest of lanes. Following this approach, a number of Emaps have been created in France, Germany, and Sweden. The experiments carried out in these test sites with the Emaps show the capability of our Emap definition to assist with the determination of the vehicle position at the lane level. Details of the processes of extraction and connection of the road segments are given in the core of this paper, as well as a discussion of the elaboration process and future guidelines in the conclusion.

Fusing GNSS, Dead-Reckoning, and Enhanced Maps for Road Vehicle Lane-Level Navigation

Nowadays, it is common that road vehicle navigation systems employ maps to represent the vehicle positions in a local reference. The most usual process to do that consists in the estimation of the vehicle positioning by fusing the Global Navigation Satellite System (GNSS) and some other aiding sensors data, and the subsequent projection of these values on the map by applying map-matching techniques. However, it is possible to benefit from map information also during the process of fusing data for positioning. This paper presents an algorithm for lane-level road vehicle navigation that integrates GNSS, dead-reckoning (odometry and gyro), and map data in the fusion process. Additionally, the proposed method brings some benefits for map-matching at lane level because, on the one hand, it allows the tracking of multiple hypothesis and on the other hand, it provides probability values of lane occupancy for each candidate segment. To do this, a new paradigm that describes lanes as piece-wise sets of clothoids was applied in the elaboration of an enhanced map (Emap). Experimental results in real complex scenarios with multiple lanes show the suitability of the proposed algorithm for the problem under consideration, presenting better results than some state-of-the-art methods of the literature.

IMM-Based Lane-Change Prediction in Highways With Low-Cost GPS/INS

The prediction of lane changes has been proven to be useful for collision avoidance support in road vehicles. This paper proposes an interactive multiple model (IMM)-based method for predicting lane changes in highways. The sensor unit consists of a set of low-cost Global Positioning System/inertial measurement unit (GPS/IMU) sensors and an odometry captor for collecting velocity measurements. Extended Kalman filters (EKFs) running in parallel and integrated by an IMM-based algorithm provide positioning and maneuver predictions to the user. The maneuver states Change Lane (CL) and Keep Lane (KL) are defined by two models that describe different dynamics. Different model sets have been studied to meet the needs of the IMM-based algorithm. Real trials in highway scenarios show the capability of the system to predict lane changes in straight and curved road stretches with very short latency times.

Making an Enhanced Map for Lane Location Based Services

Latest investigations show the benefits of fusing map information with GNSS (Global Navigation Satellite System) and dead-reckoning measurements for road vehicle navigation. However, to achieve success enhanced maps (Emaps) that take into account this new capability must be developed. In this paper, we present a method to create an Emap that includes road shape information, capable to serve as an input of a combined Fusion/Map-Matching algorithm. Details of the processes of road segments extraction and connection are given in the text. The analysis of the results obtained using the proposed method and future guidelines conclude the paper.

GPS multipath detection and exclusion with elevation-enhanced maps

The reflections of satellite signals in the environment of Global Positioning Systems (GPS) receivers cause significant errors in their position estimates. Particularly critical are the errors due to the so called non-line-of-sight (NLOS) satellites. In a NLOS situation, the only way the satellite signal reaches the receiver is by means of reflections on plane surfaces, typically buildings, causing overestimates of the pseudoranges between the satellites and the antenna. Receivers not always distinguish between the direct signals and the multipath effects, leading to un-modelled GPS errors. This paper presents a solution to the problem of multipath effects in urban areas, by means of simple elevation models of the environment. The description of the buildings is stored in a elevation-enhanced digital map (EEMap) that can be consulted to decide whether a certain satellite may be in direct view or not. The validity of the concept is proven by means of real experiments in built-up areas of Spain.

Maneuver Prediction for Road Vehicles Based on a Neuro-Fuzzy Architecture With a Low-Cost Navigation Unit

Collision avoidance is currently one of the main research areas in road intelligent transportation systems. Among the different possibilities available in the literature, the prediction of abrupt maneuvers has been shown to be useful in reducing the possibility of collisions. A supervised version of dynamic Fuzzy Adaptive System ART-based (dFasArt), which is a neuronal-architecture-based method that employs dynamic activation functions determined by fuzzy sets, is used for maneuver predicting and solving the problem of intervehicle collisions on roads. In this paper, it is shown how the dynamic character of dFasArt minimizes problems caused by noise in the sensors and provides stability on the predicted maneuvers. Several experiments with real data were carried out, and the SdFasArt results were compared with those achieved by an implementation of the Incremental Hierarchical Discriminant Regression (IHDR)-based method, showing the suitability of SdFasArt for maneuver prediction of road vehicles.

Evaluation of stereo correspondence algorithms and their implementation on FPGA

The accuracy of stereo vision has been considerably improved in the last decade, but real-time stereo matching is still a challenge for embedded systems where the limited resources do not permit fast operation of sophisticated approaches. This work presents an evaluation of area-based algorithms used for calculating distance in stereoscopic vision systems, their hardware architectures for implementation on FPGA and the cost of their accuracies in terms of FPGA hardware resources. The results show the trade-off between the quality of such maps and the hardware resources which each solution demands, so they serve as a guide for implementing stereo correspondence algorithms in real-time processing systems.