Recognizing distractions on the road is crucial to reduce traffic accidents. Video-based networks are typically used, but are limited by their computational cost and are vulnerable to viewpoint changes. In this paper, we propose a novel approach for pose-based driver action classification using speech recognition networks, which is lighter and more viewpoint invariant that video-based one. We leverage the similarity in the encoding of information between audio and pose data, representing poses as key points over time. Our architecture is based on Squeezeformer, an efficient attentionbased speech recognition network. We introduce a selection of data augmentation techniques to enhance generalization. Experiments on the Drive&Act dataset demonstrate superior performance compared to state-of-the-art methods. Additionally, we explore the integration of object information and the impact of viewpoint changes. Our results highlight the effectiveness and robustness of speech recognition networks in pose-based action classification.

Explaining the decision made by end-to-end autonomous driving is a difficult task. These approaches take raw sensor data and compute the decision as a black box with large deep learning models. Understanding the output of deep learning is a complex challenge due to the complicated nature of explainability; as data passes through the network, it becomes untraceable, making it difficult to understand. Explainability increases confidence in the decision by making the black box that drives the vehicle transparent to the user inside. Achieving a Level 5 autonomous vehicle necessitates the resolution of that challenging task. In this work, we propose a model that leverages the driver’s attention to obtain explainable decisions based on an attention map and the scene context. Our novel architecture addresses the task of obtaining a decision and its explanation from a single RGB sequence of the driving scene ahead. We base this architecture on the Transformer architecture with some efficiency tricks in order to use it at a reasonable frame rate. Moreover, we integrate in this proposal our previous ARAGAN model, which obtains SOTA attention maps, to improve the performance of the model thanks to understand the sequence as a human does. We train and validate our proposal on the BDD-OIA dataset, achieving on-pair results or even better than other state-of-the-art methods. Additionally, we present a simulation-based proof of concept demonstrating the model’s performance as a copilot in a close-loop vehicle to driver interaction.

Autonomous vehicle (AV) is one of the most challenging engineering tasks of our era. High-Definition (HD) maps are a fundamental tool in the development of AVs, being considered as pseudo sensors that provide a trusted baseline that other sensors cannot. Our approach is focused on the use of OpenDRIVE standard based HD maps in order to conduct the different mapping and planning tasks involved in Autonomous Driving (AD). In this paper we present a method for exploiting the HD map potential for two specific purposes: i) Global Path Planning and ii) Monitoring the relevant lanes and regulatory elements around the ego-vehicle to support the perception module. Mapping and planning modules are connected to the other modules of the AV stack by using ROS (Robot Operating System). Our AD architecture has been validated both in local and CARLA Autonomous Driving Leaderboard cloud, where we can appreciate a considerable improvement in the metrics by incorporating information from the HD map, not only used to conduct the Global Path Planning task but also providing prior information to the Perception module. Code is available in https://github.com/AlejandroDiazD/opendrive-mapping-planning.

The implementation of Autonomous Driving stacks (ADS) is one of the most challenging engineering tasks of our era. Autonomous Vehicles (AVs) are expected to be driven in highly dynamic environments with a reliability greater than human beings and full autonomy. Furthermore, one of the most important topics is the way to democratize and accelerate the development and research of holistic validation to ensure the robustness of the vehicle. In this paper we present a powerful ROS (Robot Operating System) based modular ADS that achieves state-of-the-art results in challenging scenarios based on the CARLA (Car Learning to Act) simulator, outperforming several strong baselines in a novel evaluation setting which involves non-trivial traffic scenarios and adverse environmental conditions (Qualitative results). Our proposal ranks in second position in the CARLA Autonomous Driving Leaderboard (Map Track) and gets the best score considering modular pipelines, as a preliminary stage before implementing it in our real-world autonomous electric car. To encourage the use research in holistic development and testing, our code is publicly available at https://github.com/RobeSafe-UAH/CARLA_Leaderboard.

Abstract In this work, we present a complete system to produce an automatic linguistic reporting about the customer activity patterns inside open malls, a mixed distribution of classical malls joined with the shops on the street. These reports can assist to design marketing campaigns by means of identifying the best places to catch the attention of customers. Activity patterns are estimated with process mining techniques and the key information of localization. Localization is obtained with a parallelized solution based on WiFi fingerprint system to speed up the solution. In agreement with the best practices for human evaluation of natural language generation systems, the linguistic quality of the generated report was evaluated by 41 experts who filled in an online questionnaire. Results are encouraging, since the average global score of the linguistic quality dimension is 6.17 (0.76 of standard deviation) in a 7-point Likert scale. This expresses a high degree of satisfaction of the generated reports and validates the adequacy of automatic natural language textual reports as a complementary tool to process model visualization.

Localization and mapping in indoor environments, such as airports and hospitals, are key tasks for almost every robotic platform. Some researchers suggest the use of RO (Range Only) sensors based on WiFi (Wireless Fidelity) technology with SLAM (Simultaneous Localization And Mapping) techniques. The current state of the art in RO SLAM is mainly focused on the filtering approach, while the study of smoothing approach with RO sensors is quite incomplete. This paper presents a comparison between a filtering algorithm, the EKF, and a smoothing algorithm, the SAM (Smoothing And Mapping). Experimental results are obtained, first in an outdoor environment using two types of RO sensors and then in an indoor environment with WiFi sensors. The results demonstrate the feasibility of the smoothing approach with WiFi sensors in indoors.

This paper presents the results of a set of extensive experiments carried out under both daytime and nighttime real traffic conditions. The data were captured using an enhanced or extended Floating Car Data system (xFCD) that includes a stereo vision sensor for detecting the local traffic ahead. The collected information is then used to propose a novel approach to the level-of-service (LOS) calculation. This calculation uses information from both the xFCD and the magnetic loops deployed in the infrastructure to construct a speed/occupancy hybrid plane that characterizes the traffic state of a continuous route. In the xFCD system, the detection component implies the use of previously developed monocular approaches in combination with new stereo vision algorithms that add robustness to the detection and increase the accuracy of the measurements corresponding to relative distance and speed. In addition to the stereo pair of cameras, the vehicle is equipped with a low-cost Global Positioning System (GPS) and an electronic device for controller-area-network bus interfacing. The xFCD system has been tested in a 198-min sequence recorded in real traffic scenarios under different weather and illumination conditions. The results are promising and demonstrate that the xFCD system is ready for being used as a source of traffic status information. As an indicative example of the developed xFCD system, we construct a novel route LOS calculation that combines hybrid information about speed and occupancy from both the xFCD system and the magnetic loops in the infrastructure.

This paper presents the results of a set of extensive experiments carried out in daytime and nighttime conditions in real traffic using an enhanced or extended Floating Car Data system (xFCD) that includes a stereo vision sensor for detecting the local traffic ahead. The detection component implies the use of previously monocular approaches developed by our group in combination with new stereo vision algorithms that add robustness to the detection and increase the accuracy of the measurements corresponding to relative distance and speed. Besides the stereo pair of cameras, the vehicle is equipped with a low-cost GPS and an electronic device for CAN Bus interfacing. The xFCD system has been tested in a 198-minutes sequence recorded in real traffic scenarios with different weather and illumination conditions, which represents the main contribution of this paper. The results are promising and demonstrate that the system is ready for being used as a source of traffic state information.

This work presents a general framework for people indoor activity recognition. Firstly, a Wireless Fidelity (WiFi) localization system implemented as a Fuzzy Rule-based Classifier (FRBC) is used to obtain an approximate position at the level of discrete zones (office, corridor, meeting room, etc). Secondly, a Fuzzy Finite State Machine (FFSM) is used for human body posture recognition (seated, standing upright or walking). Finally, another FFSM combines both WiFi localization and posture recognition to obtain a robust, reliable, and easily understandable activity recognition system (working in the desk room, crossing the corridor, having a meeting, etc). Each user carries with a personal digital agenda (PDA) or smart-phone equipped with a WiFi interface for localization task and accelerometers for posture recognition. Our approach does not require adding new hardware to the experimental environment. It relies on the WiFi access points (APs) widely available in most public and private buildings. We include a practical experimentation where good results were achieved.