Driver Assistance Functions Research Articles

Development of an electronic differential system for electric vehicles based on deep neural network

Background. Modern cars typically use a transmission system where an internal combustion engine transmits force through a gearbox to a differential shaft, which then transmits torque to the wheels. In electric vehicles, however, electric motors can be attached directly to the wheels through a reduction gearbox. With a right control system, these vehicles can apply different torque to the wheels on each side of the vehicle, significantly enhancing handling. This set-up is known as an electronic differential. Its implementation can vary from simple, using the Ackerman–Jeantand model, which cannot fully replace a simple mechanical differential, to complex systems employing robust control mechanisms with yaw torque. The latter method requires a rather complex control system and expensive sensors. A third –innovative approach involves using neural networks to control electric vehicle speeds. Aim. The aim is to provide materials and software for implementing an electronic differential using artificial neural networks to expand the solutions available for controlling multiple electric derives in various electric vehicles. Materials and Methods. The neural network considers multiple factors affecting the operation of electronic differential by selecting special coefficients, weights, to neurons within the network. As a basis for obtaining data, special route maps based on the transition curve will be used. Results. Testing yielded a prediction accuracy of 0.7273, with a standard deviation of 0.064 for the training set and 0.065 for the testing set. Conclusion. Employing neural networks for controlling electric drives is a promising alternative to traditional algorithms. Owing to their flexibility, these neural networks can also implement additional driver assistance functions, such as ESP, ABS, cruise control, etc.

Investigating Micro-Driving Behavior of Combined Horizontal and Vertical Curves Using an RF Model and SHAP Analysis

The free-flowing traffic environment of the freeway is an important application scenario for automatic driving. In this scenario, the freeway’s geometric design is an important factor because no other vehicle affects the driving process of the target vehicle. The freeway’s combined curves have more safety problems, but there are no quantitative guidelines for their geometric design. They present more challenges for automatic driving or driver assistance functions. If the relationship between human-drivers’ micro-behavior and the geometric design of combined curves is examined, it could provide theoretical support for the enhancement of automated driving and driver assistance functions as well as the quantitative design of combined curves. The paper analyzed the speed change and lane departure behaviors of combined curves, considering downslope curves, upslope curves, sag curves, and crest curves. The relationship between micro-driving behaviors and combined curves’ geometric design were determined using random forest models. The SHAP values of each variable were calculated. The results showed that (1) on a downslope curve and sag curve the speed change behavior should be paid more attention; on an upslope curve and crest curve, the lane departure behavior should be paid more attention; (2) the priority of geometric design parameters for four types of combined curves were different. Based on the results, drivers and autonomous vehicles can pay different levels of attention to their speed change and departure behavior on different combination curves, and take targeted improvement measures in time according to the driving status of the vehicles. Road designers can also prioritize more important road design parameters in the design process to avoid serious accidents caused by excessive speed changes and departures.

Model-Based Estimation of Vehicle Center of Gravity Height and Load

AbstractWith an increasing number of driver assistance functions and the upcoming trend of autonomous driving, knowledge of the current state and changeable parameters of the vehicle becomes more and more important. One particularly significant parameter with regard to vehicle dynamics is the center of gravity (COG) height, which mainly accounts for its roll dynamics in combination with the mass of the vehicle. Thus, a highly increased vehicle mass and COG height might lead to rollover during cornering, which underlines the need for accurate knowledge of the load of the vehicle. Based on that, an improved rollover prevention could be implemented, for instance, by enhancing the electronic stability program (ESP) of the vehicle. Therefore, the main contribution of this work is the model-based online estimation of the additional load of the vehicle, comprising its COG position and mass. This is achieved by applying a joint extended Kalman Filter (EKF) for the simultaneous state and parameter estimation. Based on a nonlinear model with roll, pitch, and vertical dynamics, an accurate and reliable estimation is possible. One major novelty of this work is the consideration of air suspension systems on top of conventional steel spring suspension systems. Therefore, a nonlinear air spring model with sufficient complexity is proposed, making it suitable for real-time applications. Further, a system theoretical observability analysis allows for an online adaptation of the Kalman Filter weights in order to account for different driving situations. The proposed estimation method is tested and validated by considering a wide range of driving situations, considering distinct loading conditions on both a test track and public roads. The estimation accuracy lies within roughly 50 kg and 1.5 cm for the vehicle mass and COG height, respectively.

Optimizing vehicle dynamics co-simulation performance by introducing mesoscopic traffic simulation

Microscopic traffic simulation is often used in conjunction with vehicle dynamics simulation to test cooperative or perception-based driver assistance functions. On the other hand, visualization and the interaction of a large number of swarm vehicles is computationally burdensome, limiting the size of scenarios. To remedy this problem, this paper introduces a mesoscopic traffic model, namely the extension of the shockwave profile model to road networks, that handles traffic as continuous flows of vehicles. Adopting the idea of level of detail from computer graphics, the modeling of swarm vehicles is carried out in less detail farther away from the EGO vehicle. These levels are defined using the classical (macroscopic, mesoscopic, and microscopic) categorization of traffic modeling. The macroscopic traffic model is responsible for the traffic demand and traffic assignment. The proposed mesoscopic model is capable of capturing the fluctuating nature of traffic on lane level. Closer to the EGO vehicle microscopic traffic simulation is employed while the EGO vehicle is modeled in full-detail including vehicle dynamics too. The 3D rendering of the simulation is performed by the vehicle dynamics simulator. The challenge in the proposed methodology is transitioning between the mesoscopic and the microscopic models, i.e., selecting the boundary and spawning/destroying vehicle agents. The paper addresses this challenge with a linear (w.r.t. the vehicle number) time complexity algorithm. Practically, a dynamic downscaling of the microscopic simulation to mesoscopic level is realized outside the vicinity of the EGO vehicle. The proposed methodology is generic, and can be adapted to most existing vehicle dynamics and microscopic traffic simulator software. The solution is tested with SUMO as microsimulation and Carla as vehicle dynamics simulation through a simple path following test case in two scenarios: a congested residential area and a complex scenario with both a highway section and an urban area. Simulation results suggest that the simulation performance could be improved by 200–500% while retaining modeling accuracy, compared to the case when only microscopic simulation is used.

A Beta-less Approach for Vehicle Cornering Stiffness Estimation

With an increasing number of driver assistance functions and the upcoming trend of autonomous driving, knowledge of the current state and changeable parameters of the vehicle becomes increasingly essential. Particularly significant parameters with regard to vehicle lateral dynamics are the cornering stiffnesses, which describe the tire lateral forces depending on the current slip angles. Thus, the cornering stiffness characterizes the current lateral force of a tire, which heavily varies over different tires and decreases by tire wear. However, its knowledge is crucial for describing the vehicle dynamics behavior. Therefore, vehicle control algorithms and driver assistance functions would highly benefit from knowledge of the vehicle cornering stiffness values, which could increase driving comfort and safety. For that purpose, this paper introduces an online estimation algorithm for the tire cornering stiffness values. Only typical vehicle series sensors are required for the algorithm, which ensures its applicability for series vehicles. Finally, vehicle measurements are conducted in order to validate the proposed estimation algorithm by means of the resulting cornering stiffnesses.

OVERVIEW OF TEST INFRASTRUCTURE INVESTMENTS ON OPEN ROADS TO TEST CONNECTED AND AUTOMATED VEHICLES

Open road testing is an essential stage of the process of approving driver assistance and other self-driving functions. However several countries have opened their network with or without limitations for autonomous vehicle testing, there are some sections where the infrastructure was designed to support testing. The aim of this article is to give an overview of such open road testing facilities around the world, introducing the main purposes, the used infrastructure (eg. sensor network, communication) and features (eg. data processing, and control functions), and compare them to the planned investments in Hungary.

Anomaly detection as vision-based obstacle detection for vehicle automation in industrial environment

Nowadays, produced cars are equipped with mechatronical actuators as well as with a wide range of sensors in order to realize driver assistance functions. These components could enable cars’ automation at low speeds on company premises, although autonomous driving in public traffic is still facing technical and legal challenges. For automating vehicles in an industrial environment a reliable obstacle detection system is required. State-of-the-art solution for protective devices in Automated Guided Vehicles is the distance measuring laser scanner. Since laser scanners are not basic equipment of today’s cars in contrast to monocameras mounted behind the windscreen, we develop a computer vision algorithm that is able to detect obstacles in camera images reliably. Therefore, we make use of our well-known operational design domain by teaching an anomaly detection how the vehicle path should look like. The result is an anomaly detection algorithm that consists of a pre-trained feature extractor and a shallow classifier, modelling the probability of occurrence. We record a data set of a real industrial environment and show a robust classifier after training the algorithm with images of only one run. The performance as an obstacle detection is on par with a semantic segmentation, but requires a fraction of the training data and no labeling.

Driving Robot for Reproducible Testing: A Novel Combination of Pedal and Steering Robot on a Steerable Vehicle Test Bench

Shorter development times, increased standards for vehicle emissions and a greater number of vehicle variants result in a higher level of complexity in the vehicle development process. Efficient development of powertrain and driver assistance functions under comparable and reproducible operating conditions is possible on vehicle test benches. Yet, the realistic simulation of real driving environments on test benches is a challenge. Current test procedures and new technologies, such as Real Driving Emission tests and Autonomous Driving, require a reproducible and even more detailed simulation of the driving environment. Due to this, the simulation of curve driving in particular is gaining in importance. This results from its significant influence on energy consumption and Autonomous Driving functions with lateral guidance, such as lane departure and evasion assistance. Reproducibility can be additionally increased by using a driving robot. At today’s vehicle test benches, pedal and shift robots are predominantly used for longitudinal dynamic tests in the performed test procedures. In order to meet these new test automation requirements for vehicle test benches, the cooperative operation of pedal and steering robots is needed on a test bench setup suitable for this purpose. In this publication, the authors present the setup of a vehicle test bench to be used in automated and reproducible vehicle-in-the-loop tests during steering events. The focus is on the test-bench-specific setup with steerable front wheels, the actuators for simulating the wheel steering torque around the steering axle and the robots used for pedals and steering wheel. Results from various test series are presented and the potential of the novel test environment is shown. The results are reproducible in various test series due to the closed-loop operation without human driving influences at the test bench.

Velocity Estimation via Wheel Circumference Identification

The article presents a velocity estimation algorithm through the wheel encoder-based odometry and wheel circumference identification. The motivation of the paper is that a proper model can improve the motion estimation in poor sensor performance cases. For example, when the GNSS signals are unavailable, or when the vision-based methods are incorrect due to the insufficient number of features, furthermore, when the IMU-based method fails due to the lack of frequent accelerations. In these situations, the wheel encoders can be an appropriate choice for state estimation. However, this type of estimation suffers from parameter uncertainty. In the paper, a wheel circumference identification is proposed to improve the velocity estimation. The algorithm listens to the incoming sensor measurements and estimates the wheel circumferences recursively with a nonlinear least squares method. The experimental results demonstrate that with the application of the identified parameters in the wheel odometry model, accurate velocity estimation can be obtained with high frequency. Thus, the presented algorithm can improve the motion estimation in the driver assistant functions of autonomous vehicles.

Navigation system with STV driver assistance function

The article deals with the problems of application of modern driver assistance systems on single-track vehicles (STV). The design features of the STV that prevent the installation of such systems are presented. Variants of a motorcycle entering a turn are considered, with concerning the gyroscopic moment of the STV wheels and the causes of situations in the turn. A new navigation system with a driver assistance function with an analysis of its operation and advantages is proposed.

Development of Precise Point Positioning Algorithm to Support Advanced Driver Assistant Functions for Inland Vessel Navigation

Development of Precise Point Positioning Algorithm to Support Advanced Driver Assistant Functions for Inland Vessel Navigation

Automated driving for car manufacturers’ vehicle logistics

Abstract Technical and legal challenges cause the implementation of Autonomous Driving in road traffic to still be a long way off. However, the introduction of driver assistance functions enables cars’ automation for low speeds already nowadays. The concept of Autonomous Transport (AT) combines automated driving with Automated Guided Vehicle’s technology. In this paper, we assess risks that emanate from AT and show fields of action for its implementation with respect to the standards for functional safety. We set up requirements for the reliability of cars’ electric power supply, actuators and sensors. Concepts for their cost-efficient fulfillment are derived. The realization of collision avoidance and navigation without additional attachments is discussed.

Development of ADAS perception applications in ROS and "Software-In-the-Loop" validation with CARLA simulator

Higher levels of autonomous driving are bringing sophisticated requirements and unpredicted challenges. In order to solve these problems, the set of functionalities in modern vehicles is growing in terms of algorithmic complexity and required hardware. The risk of testing implemented solutions in real world is high, expensive and time consuming. This is the reason for virtual automotive simulation tools for testing are heavily acclaimed. Original Equipment Manufacturers (OEMs) use these tools to create closed sense, compute, act loop to have realistic testing scenarios. Production software is tested against simulated sensing data. Based on these inputs a set of actions is produced and simulated which generates consequences that are evaluated. This creates a possibility for OEMs to minimize design errors and optimize costs of the vehicle production before any physical prototypes are produced. This paper presents the development of simple C++/Python perception applications that can be used in driver assistance functionalities. Using ROS as a prototyping platform these applications are validated and tested with "Software-In-the Loop" (SIL) method. CARLA simulator is used as a generator for input data and output commands of the autonomous platform are executed as simulated actions within simulator. Validation is done by connecting Autoware autonomous platform with CARLA simulator in order to test against various scenes in which applications are applicable. Vision based lane detection, which is one of the prototypes, is also tested in a real world scenario to demonstrate the applicability of algorithms developed with simulators to real-time processing

Enabling Assistance Functions for the Safe Navigation of Inland Waterways

Inland navigation and shipping are important pillars of the European Transport System. To support the skipper during safety-critical operations, precise position, navigation, and timing (PNT) data are required. This work discusses the role of PNT information for enabling inland waterway navigation-assistance functions, such as bridge-collision warning, mooring aiding, and automatic guidance. The real-time kinematic technique is developed to provide integrity information alongside reliable and centimeter-level-accurate PNT data. In addition, the transmission of Global Navigation Satellite Systems (GNSS) correction data is investigated. Since GSM does not currently meet the availability and stability communication requirements along inland waterways, use of the automatic identification system (AIS) for data transmission is explored. The proposed navigation solution and the communication developments were analyzed in real time in challenging GNSS signal-degraded scenarios on the authorized inland waterway testbed. Despite the further developments required for the AIS communication infrastructure, it is shown that our system architecture can nearly meet the integrity and accuracy requirements for driver-assistance functions on inland waterways.

Optimal Control of Overtaking Maneuver for Intelligent Vehicles

In the paper a hierarchical overtaking strategy, which is a driver assistance function or rather an autonomous function in electric/autonomous vehicles, is proposed. The solution uses speed and acceleration signals from the surrounding vehicles. These signals are processed with clustering methods in order to achieve probability density functions and predict their expected motion. The strategy includes several additional layers, such as decision making concerning the maneuver, the computation of the required trajectory, and the tracking control of the vehicle. Trajectory generation is formed as an optimization task, which is able to include the prediction model of the surrounding vehicles in the constraints. A robust Linear Parameter Varying (LPV) control design method is proposed to guarantee the tracking of the computed reference. The proposed strategy is able to guarantee the safe motion of the vehicles and handle the interactions with the other traffic participants.

DATA PROCESSING AND RECORDING USING A VERSATILE MULTI-SENSOR VEHICLE

Abstract. In this paper we present a versatile multi-sensor vehicle which is used in several research projects. The vehicle is equipped with various sensors in order to cover the needs of different research projects in the area of object detection and tracking, mobile mapping and change detection. We show an example for the capabilities of this vehicle by presenting camera- and LiDAR-based pedestrian detection methods. Besides this specific use case, we provide a more general in-depth description of the vehicle’s hard- and software design and its data-processing capabilities. The vehicle can be used as a sensor carrier for mobile mapping, but it also offers hardware and software components to allow for an adaptable onboard processing. This enables the development and testing of methods related to real-time applications or high-level driver assistance functions. The vehicle’s hardware and software layout result from several years of experience, and our lessons learned can help other researchers set up their own experimental platform.

Proposal of a new virtual evaluation approach of preventive safety applications and advanced driver assistance functions – application: AEB system

This study presents a new virtual evaluation approach of preventive safety applications and advanced driver assistance functions. The approach identifies the worst-case scenarios for a given advanced driver assistance function, AEB system in this study, based on field operational tests (FOT) [safety pilot model deployment (SPMD), in this study]. The authors begin with a description of the studied AEB system and a synthesis of the most relevant tests scenarios. Then, they model the distribution of each test parameter retrieved from the SPMD database by applying two estimation methods (kernel method and expectation-maximisation algorithm). A comparison was made between the two methods to choose the best one. These distributions are then sampled using the proposed sampling strategy based on Metropolis-Hastings algorithm. Then, the idea is to take the samples of each parameter retrieved with this sampler, simulate them on a vehicular software simulator (PreScan) and to get their simulation results. For each test and in case of impact, a proportional score to the speed of impact reduction is attributed. Finally, a risk classification is done based on the scoring results which allows to recover high and very high-risk cases to build a set of worst-case scenarios.

Cross‐regional driver–vehicle interaction design: an interview study on driving risk perceptions, decisions, and ADAS function preferences

A cross-regional study of driving behaviour and technological preferences in typical driving scenarios, especially dangerous pre-crash scenarios, is presented as a contribution to the user experience design of in-vehicle driver assistance functions. Data from 46 participants are collected by one-to-one interviews following viewing of 11 video clips previously obtained from naturalistic field operational tests and representative of typical real driving scenarios. Six questions relating to each driving scenario are asked to reveal the differences between Chinese drivers and Swedish drivers. The results show similarities and differences in driving risk perceptions, decisions, and preferences concerning the assistance and specifics of potential advanced driver assistance system (ADAS) functions of drivers in China and Sweden. The preferences for assistance and ADAS functions are found to be correlated with relative driving risk perceptions and decisions in typical driving scenarios for both country groups. Based on the results, some suggestions for the design of driver–vehicle interactions for Chinese drivers are presented.

Probabilistic Threat Assessment with Environment Description and Rule-based Multi-Traffic Prediction for Integrated Risk Management System

The objective of this paper is to propose an original probabilistic threat assessment method to predict and avoid all possible kinds of collision in multi-vehicle traffics. The main concerns in risk assessment can be summarized as three requirements: 1) a description of a traffic situation containing the geometric description of the road, dynamic and static obstacle tracking, 2) a prediction of multiple traffics' reachable set under the reasonable behavior restriction, and 3) an assessment of collision risk which corresponds with driver sensitivity and can be applied to many complex situations without loss of generality. To fulfill these three requirements, the proposed algorithm for estimating the probability of collision occurrence of the ego vehicle follows the basic idea of the particle filtering and the collision probability can be numerically implemented and calculated. The overall performance of the proposed threat assessment algorithm is verified via vehicle tests in real road. It has been shown that the threat assessment performance for the given driving situations can be significantly enhanced by the proposed algorithm. And this enhancement of risk assessment performance led to capabilities improvement of driver assistance functions of ADASs.

Comparison of fail-operational software architectures from the viewpoint of an automotive application

Due to the trend towards advanced driver assistance functions and fully automated driving, many future automotive systems have to provide fail-operational behaviour, maintaining certain functions for a certain time even in case of certain faults. Such behaviour is well-known in other domains where the availability of certain functions is important due to safety or financial reasons. The ability to tolerate faults and failures in systems, hardware and software therefore becomes increasingly important. Since there are topologies already available to cope with these new requirements, it is the task of the automotive industry to select the ones that fit to its specific constraints. These constraints include a high cost pressure, the scarcity of packaging space, limited resources for software, and requirements of the ISO 26262:2011 (i.e. the functional safety standard in the automotive industry). In order to support the decision on selecting the right software architecture, the strengths and weaknesses of topologies used by other industrial sectors have to be known thoroughly. This paper investigates three typical fault tolerant software architectures by use of a structured analysis technique, and by applying a set of criteria specific for the automotive domain. The intention of this paper is twofold: The primary goal is to gain an understanding related to the properties of the compared architectures. The second goal is to prove that the chosen software architecture comparison method is suitable to compare schematic, high level topologies.