ADAS USING TOUCH SENSOR

Advanced Driver Assistance Systems (ADAS) have emerged as a critical technology in enhancing vehicle safety, providing convenience to drivers, and facilitating the transition toward fully autonomous vehicles. The core of these systems relies heavily on computer vision (CV) technologies, which enable vehicles to interpret and understand their surrounding environment in real time. The integration of computer vision in ADAS allows vehicles to recognize and process various objects such as pedestrians, vehicles, traffic signs, and lane markings, significantly improving both safety and the driving experience. In this paper, we explore the pivotal role of computer vision in the development and implementation of ADAS in autonomous vehicles. We begin by reviewing the evolution of ADAS, detailing the different sub-systems such as lane departure warning, automatic emergency braking, adaptive cruise control, and parking assistance, all of which depend on sophisticated image processing techniques. Furthermore, the paper examines the intersection of computer vision with other emerging technologies such as machine learning, deep learning, and sensor fusion. We discuss how these innovations enable ADAS to function more effectively in complex, dynamic environments. The methodology section focuses on cutting-edge algorithms used for object detection, semantic segmentation, and real-time image processing, which are integral to the functionality of modern ADAS. Additionally, we evaluate the challenges that computer vision systems face in real-world applications, including issues with sensor calibration, environmental conditions, computational limitations, and the integration of data from multiple sensors (e.g., cameras, LiDAR, radar). The paper also presents an analysis of the results obtained from various real-world implementations of ADAS technologies, highlighting improvements in driver safety and the reduction of traffic-related accidents. We explore the potential of computer vision in enabling fully autonomous driving by addressing the current gaps in technology and the advancements needed to close these gaps. Finally, the paper concludes with a discussion on the future of ADAS and the role of computer vision in realizing fully autonomous, reliable, and efficient transportation systems. We argue that continued research and development in computer vision, combined with innovations in deep learning and sensor technologies, are critical to advancing the capabilities and reliability of autonomous vehicles in the near future.

The integration of artificial intelligence (AI) in the automotive industry is revolutionizing various facets of vehicle design, manufacturing, and operation. This paper explores the key areas where AI plays a pivotal role, focusing on autonomous driving, advanced driver assistance systems (ADAS), predictive maintenance, and in-vehicle AI systems. Autonomous driving technologies leverage AI for perception, decision-making, and control, pushing the boundaries toward full vehicle autonomy. ADAS features such as lane-keeping assistance, adaptive cruise control, and emergency braking enhance driver safety by using real-time AI analysis. Furthermore, predictive maintenance systems utilize AI to forecast vehicle component failures, reducing operational downtime and extending vehicle lifespans. In manufacturing, AI improves quality control and supply chain efficiency, while in-vehicle AI systems offer personalized user experiences through voice recognition and augmented reality interfaces. Lastly, AI’s role in smart traffic systems and the sustainability of electric vehicles (EVs) is discussed, highlighting its impact on energy efficiency and traffic optimization. This paper emphasizes how AI-driven advancements are transforming the automotive sector, enhancing safety, efficiency, and sustainability.

• The performance of Adaptive Cruise Control (ACC) is limited on road curves. • The Intelligent Speed Adaptation for Visibility (V-ISA) adjusts speed based on road conditions. • A driving simulation study assessed the influence of ACC and V-ISA on driver behaviour. • Integrating ACC and V-ISA significantly improved longitudinal driving performance. • The system achieved high system acceptance with no added mental workload. Advanced Driver Assistance Systems (ADAS) have become an integral part of modern vehicles, significantly changing the driving experience. Adaptive Cruise Control (ACC) assists drivers in maintaining a constant speed and a safe following distance; however, it does not moderate the speed along curves when sight limitations occur, thus compromising safety. To address these challenges, this study integrates the Intelligent Speed Adaptation for Visibility (V-ISA) system with ACC. The V-ISA system dynamically adjusts the vehicle speed based on real-time visibility conditions, thereby enhancing safety on road curves where sight distance is restricted. Additionally, this study evaluates the impact of the ACC + V-ISA system on driver performance, specifically focusing on speed management, lateral vehicle control, and driver experience with assistance systems. The study was conducted in a simulated environment with forty-five participants. The results revealed that the ACC + V-ISA system effectively reduced vehicle speed on curves with limited visibility. The ability of the system to regulate speed according to prevailing road conditions highlights its potential to improve road safety. Moreover, the findings suggest that using the V-ISA system integrated with ACC maintains user acceptance and satisfaction while imposing no additional mental workload compared to using ACC alone. These insights are crucial for the future design of ADAS and autonomous vehicles and emphasize the importance of introducing this system into vehicles. LED bar colour

This thesis deals with the analysis and understanding of drivers’ behaviours under car-following. The aim is to enhance the modelling tools toward the development of new ADAS (Advanced Driving Assistance System) logics, characterized by a more human-like behaviour. After having introduced the argument of the thesis (and motivated the work) and having recalled the state of the art most relevant in the field of car-following (as well as in the instruments for observing car-following in the real world), the thesis evolves toward three main sections: actual observation of real-world data and collection of the datasets to be employed for theoretical analysis; theoretical enhancements and propositions; applications to ACC (Adaptive Cruise Control), as a relevant field for ADAS. The data employed in this work have been collected in three different field surveys, two of them carried out in Italy and the other in the United Kingdom. In all cases data have been collected by instrumented vehicles, equipped in such a way to observe and record car-following trajectories. Data have been framed into different theoretical paradigms in order to both validate each theory and to establish the links between these theories. Links have been established both in a formal way (through theoretical investigation) and in a data-driven way. The considered theoretical paradigm for modelling car-following follows different approaches: one is based on the psycho-physical approach and two others are based on an engineering-inspired approach. In particular, the considered psycho-physical approach has been the Action Point theory (Wiedemann, 1974); a revised version of the paradigm, more compliant with the original version of Barbosa (1961) and Todosoiev (1963) has been proposed and justified with reference to the collected data. The first engineering paradigm has been based on a state-space approach. The proposed approach has been shown to be consistent with the Action Point theory. The parameters of the model have been estimated by means of the collected data and the obtained results have been discussed; they are consistent with observations and justify the adopted model. The other engineering model is based on a linear approximation (at any time t, in a discrete-time approach) of the response of the follower to the leader’s stimuli. Also the linear model is shown to be a very good approximation of the observed data; moreover, it has been shown to lead to an harmonic oscillation around the desired spacing at steady-state. This oscillation is consistent with both the Action Point theory and (partially) with the proposed state-space approach. The linear model is particularly suitable for real-time ACC-oriented application; thus it is the model employed in section 4 of this work, where a fully-adaptive ACC system is developed, able to actuate a driving-style actually consistent with driver’s expectations and preferences.

Advanced rider assistance systems (ARAS) play a crucial role in enhancing motorcycle safety through features such as collision avoidance, blind-spot detection, and adaptive cruise control, which rely heavily on sensors like radar, cameras, and LiDAR. However, their performance is often compromised under adverse weather conditions, leading to sensor interference, reduced visibility, and inconsistent reliability. This study evaluates the effectiveness and limitations of ARAS technologies in rain, fog, and snow, focusing on how sensor performance, algorithms, techniques, and dataset suitability influence system reliability. A thematic analysis was conducted, selecting studies focused on ARAS in adverse weather conditions based on specific selection criteria. The analysis shows that while ARAS offers substantial safety benefits, its accuracy declines in challenging environments. Existing datasets, algorithms, and techniques were reviewed to identify the most effective options for ARAS applications. However, more comprehensive weather-resilient datasets and adaptive multi-sensor fusion approaches are still needed. Advancing in these areas will be critical to improving the robustness of ARAS and ensuring safer riding experiences across diverse environmental conditions.

Suggestions for improving the drivers’ understanding and trust in the advanced driver assistance system (ADAS) are proposed. A user survey and an in-depth interview were conducted to investigate what information the drivers should receive from the system to gain trust in ADAS. Results obtained from the user survey indicate that not only the general public but also experts lacked understanding of the system behaviour under automatic operation. In particular, drivers lacked understanding of the ‘temporary mode disengagement’ and ‘restricted range of operation’. Results obtained from the in-depth interview indicate that drivers require information on the system operation if it acts differently from the user’s expectation. The suggestions in this study were refined based on insights from the qualitative analysis. The key result was interpreted by the theoretical study of Donald Norman’s ‘seven stages action model’ and ‘mental model’. Moreover, we determined the drivers’ need for information on the status of the system and the reasons behind its automatic operations. We sorted the system’s automatic operations into ‘temporary mode disengagement’, ‘restricted range of operation’, ‘detection of the front car of adaptive cruise control (ACC)’, and ‘failure of lane detection of lane-keeping assist (LKA)’. Providing information enables drivers to understand the system’s automatic operation and general purpose. The findings in this study provide an approach to form the right mental model for drivers when employing ADAS. We expect that our findings will be of help to design ADAS user interface increasing trust and safety.

Much of the existing research on drivers’ understanding of adaptive cruise control (ACC), a type of advanced driver assistance system, was conducted several years ago. Through an online survey, this study aimed to assess ACC knowledge among ACC owners and non-owners now that this system is more widely available. Along with knowledge of ACC features and limitations, demographic information, experience with technology, and experience with ACC (for owners) were also collected to investigate which factors predicted knowledge of ACC features and limitations. Results showed that owners today may have a better understanding of some of the main limitations of ACC compared with research conducted over 10 years ago. However, a large percentage of owners still had misperceptions about their ACC system. While owners had a slightly higher percentage of correct answers overall, they did not differ from non-owners in their knowledge of limitations. As this technology is becoming more common, even non-owners may be becoming aware of common limitations; owning and using ACC does not seem to result in a better system understanding. Higher income was associated with a higher percentage of correct responses on the ACC knowledge questionnaire for both owners and non-owners, and for non-owners, higher education level was also significantly associated with a higher percentage of correct responses. Future research should focus on developing training materials that are accessible to all drivers, so that drivers in lower education and income groups are also supported to understand how advanced driver assistance systems work and benefit from these technologies.

Abstract: The development of modern driver assistance systems, such as Adaptive Cruise Control (ACC) and its enhanced version ACC+, represents a significant step towards improving road safety. This article analyzes the operation of these systems in the context of the causes and effects of traffic accidents. Based on the reconstruction of a specific traffic incident, a detailed assessment of the impact of ACC and ACC+ on traffic safety was conducted. The study is based on the analysis of measurable technical parameters that change under varying road conditions, both with the systems activated and deactivated. The paper also discusses examples of preventive strategies for similar incidents, utilizing new technologies applied in modern vehicles. Particular attention is given to differences in vehicle behavior under various operational scenarios of ACC and ACC+ systems, with a focus on their impact on the safety of road users. This article contributes to the development of knowledge on minimizing accident risks through advanced driver assistance systems. Keywords: ACC; ACC+; Safety; Traffic

Advanced Driver Assistance System (ADAS) is a very important part of an up to date vehicle. For achieving highlevel objectives in such ADAS functionality like LKA (lane keeping assistance), LDW (lane departure warning system), FCW (forward collision warning) the quality of the algorithms under the hood must be extremely high. In the last few years, it is common that these algorithms are based on DNNs (deep neural networks) applied to the tasks of semantic and instance segmentation, 2D/3D object detection and visual object tracking. Recent state-of-the-art DNN models as usual solve only one single task from the listed above and running several neural networks is rather computationally expensive and even impossible due to the lack of the GPU memory. One of the approaches used to overcome such a problem is a shared backbone (also called feature extractor or encoder). The backbone consumes most of the computing resources thus the model with a shared backbone achieves better inference performance. Unfortunately, the training procedure for a shared backbone model has several difficulties. The first one is the lack of datasets with all the required and uniform annotation types. The second problem is a more sophisticated backpropagation procedure. In this paper, we consider several methods for multi-task neural network training and present the results of such training procedures on several public datasets with dissimilar annotation types. The shared backbone is applied to the following three tasks performed simultaneously on the road scene: semantic segmentation, 2D object detection and 3D object detection. While the performance of the DNNs with shared backbone increased significantly, we obtained the quality evaluation results, which are quite close to the original separate state-of-the-art DNNs and even outperforms them in some evaluation indices.

Advanced driver assistance systems (ADAS) are designed to increase driver's situational awareness and road safety by providing essential information, warning and automatic intervention to reduce the possibility/severity of an accident. Of the various types of ADAS modalities available, camera based ADAS are being widely adopted for their usefulness in varied applications, overall reliability and adaptability to new requirements. But camera based ADAS also represents a complex, high-performance, and low-power compute problem, requiring specialized solutions. This paper introduces a high performance front camera ADAS based on a small area, low power System-on-Chip (SoC) solution from Texas Instruments called Texas Instruments Driver Assist 3x (TDA3x). The paper illustrates compute capabilities of the device in implementation of a typical front camera ADAS. The paper also introduces key programming concepts related to heterogeneous programmable compute cores in the SoC and the software framework to use those cores in order to develop the front camera solutions. These aspects will be of interest not only to the ADAS developers but for computer vision and compute intensive embedded system development.

Model predictive control policy design, solutions, and stability analysis for longitudinal vehicle control considering shockwave damping

The advanced driver assistance system (ADAS) for adaptive cruise control and collision avoidance is strongly dependent upon the robust image recognition technology such as lane detection, vehicle/pedestrian detection, and traffic sign recognition. However, the conventional ADAS cannot realize more advanced collision evasion in real environments due to the absence of intelligent vehicle/pedestrian behavior analysis. Moreover, accurate distance estimation is essential in ADAS applications and semiglobal matching (SGM) is most widely adopted for high accuracy, but its system-on-chip (SoC) implementation is difficult due to the massive external memory bandwidth. In this paper, an ADAS SoC with behavior analysis with Artificial Intelligence functions and hardware implementation of SGM is proposed. The proposed SoC has dual-mode operations of highperformance operation for intelligent ADAS with real-time SGM in D-Mode (d-mode) and ultralow-power operation for black box system in parking-mode. It features: 1) task-level pipelined SGM processor to reduce external memory bandwidth by 85.8%; 2) region-of-interest generation processor to reduce 86.2% of computation; 3) mixed-mode intention prediction engine for dualmode intelligence; and 4) dynamic voltage and frequency scaling control to save 36.2% of power in d-mode. The proposed ADAS processor achieves 862 GOPS/W energy efficiency and 31.4GOPS/ mm(2) area efficiency, which are 1.53x and 1.75x improvements than the state of the art, with 30 frames/s throughput under 720p stereo inputs.

This paper details the design of two advanced driver assistance systems from concept to fully integrated solutions that are installed and tested on a driving simulator. The systems were tuned and validated using offline simulations before they were installed on the Leeds driving simulator. The simulator is equipped with a large amplitude motion base that is able to generate realistic sustained accelerations. The work forms a part of the effects of automated systems on safety project that is funded by the EPSRC. This simulator study acts a pilot study for the project. Drivers were subjected to a range of scenarios, some of which required evasive action. The two systems, adaptive cruise control and a lane keeping system, were tested in isolation, while the performance of each system and the driver were monitored. The driver was then asked to complete questionnaires on each controller. The nature of this pilot study meant that the sample size was small, but the results still demonstrate some interesting trends. The results of assisted and non-assisted driving were compared and the merits of each system were discussed.

Nonlinear MPC for Emission Efficient Cooperative Adaptive Cruise Control

Various advanced driver assistance systems (ADASs) have recently been developed, such as Adaptive Cruise Control and Precrash Safety System. However, most ADASs can operate in only some driving situations because of the difficulty of recognizing contextual information. For closer cooperation between a driver and vehicle, the vehicle should recognize a wider range of situations, similar to that recognized by the driver, and assist the driver with appropriate timing. In this paper, we assumed a double articulation structure in driving behavior data and segmented driving behavior into meaningful chunks for driving scene recognition in a similar manner to natural language processing (NLP). A double articulation analyzer translated the driving behavior into meaningless manemes, which are the smallest units of the driving behavior just like phonemes in NLP, and from them it constructed navemes, which are meaningful chunks of driving behavior just like morphemes. As a result of this two-phase analysis, we found that driving chunks equivalent to language words were closer to the complicated or contextual driving scene segmentation produced by human recognition.