Search Results

The prediction of agents' future trajectory is a crucial task in supporting advanced driver-assistance systems (ADAS) and plays a vital role in ensuring safe decisions for autonomous driving (AD). Currently, prevailing trajectory prediction methods heavily rely on high-definition maps (HD maps) as a source of prior knowledge. While HD maps enhance the accuracy of trajectory prediction by providing information about the surrounding environment, their widespread use is limited due to their high cost and legal restrictions. Furthermore, due to object occlusion, limited field of view, and other factors, the historical trajectory of the target agent is often incomplete This limitation significantly reduces the accuracy of trajectory prediction. Therefore, this paper proposes ETSA-Pred, a mapless trajectory prediction model that incorporates enhanced temporal modeling and spatial self-attention.

SLAM (Simultaneous Localization and Mapping) plays a key role in autonomous driving. Recently, 4D Radar has attracted widespread attention because it breaks through the limitations of 3D millimeter wave radar and can simultaneously detect the distance, velocity, horizontal azimuth and elevation azimuth of the target with high resolution. However, there are few studies on 4D Radar in SLAM. In this paper, RI-FGO, a 4D Radar-Inertial SLAM method based on Factor Graph Optimization, is proposed. The RANSAC (Random Sample Consensus) method is used to eliminate the dynamic obstacle points from a single scan, and the ego-motion velocity is estimated from the static point cloud. A 4D Radar velocity factor is constructed in GTSAM to receive the estimated velocity in a single scan as a measurement and directly integrated into the factor graph. The 4D Radar point clouds of consecutive frames are matched as the odometry factor.

For safe driving function, signs must be visible. Sign visibility is function of its luminance intensity. During day, due to ambient light conditions sign luminance is not a major concern. But during night, due to absence of sun light sign board retro-reflectivity plays a crucial role in sign visibility. The vehicle headlamp color, beam pattern, lamp installation position, the relative seating position of driver and moon light conditions are important factors. Virtual simulation approach is used for analyzing the sign board visibility. Among various factors for example the headlamp installation position from ground, distance between two lamps and eye position of driver are considered for analyzing the sign board visibility in this paper. Many automotive organizations have widely varying requirements and established testing guidelines to ensure visibility of signs in head lamp physical testing but there are no guidelines during design stage for headlamp for sign visibility.

Temporal light modulation (TLM), colloquially known as “flicker,” is an issue in almost all lighting applications, due to widespread adoption of LED and OLED sources and their driving electronics. A subset of LED/OLED lighting systems delivers problematic TLM, often in specific types of residential, commercial, outdoor, and vehicular lighting. Dashboard displays, touchscreens, marker lights, taillights, daytime running lights (DRL), interior lighting, etc. frequently use pulse width modulation (PWM) circuits to achieve different luminances for different times of day and users’ visual adaptation levels. The resulting TLM waveforms and viewing conditions can result in distraction and disorientation, nausea, cognitive effects, and serious health consequences in some populations, occurring with or without the driver, passenger, or pedestrian consciously “seeing” the flicker.

Ergonomics plays an important role in automobile design to achieve optimal compatibility between occupants and vehicle components. The overall goal is to ensure that the vehicle design accommodates the target customer group, who come in varied sizes, preferences and tastes. Headroom is one such metric that not only influences accommodation rate but also conveys a visual perception on how spacious the vehicle is. An adequate headroom is necessary for a good seating comfort and a relaxed driving experience. Headroom is intensely discussed in magazine tests and one of the key deciding factors in purchasing a car. SAE J1100 defines a set of measurements and standard procedures for motor vehicle dimensions. H61, W27, W35, H35 and W38 are some of the standard dimensions that relate to headroom and head clearances.

A total of 93 tests were conducted in daytime conditions to evaluate the effect on the Time to Collision (TTC), emergency braking, and avoidance rates of the Forward Collision Warning (FCW) and Automatic Emergency Braking (AEB) provided by a 2022 Tesla Model 3 against a 4ActivePA adult static pedestrian target. Variables that were evaluated included the vehicle speed on approach, pedestrian offsets, pedestrian clothing, and user-selected FCW settings. As a part of the Tesla’s Collision Avoidance AssistTM, these user-selected FCW settings change the timing of the issuance of the visual and/or audible warning provided. This testing evaluated the Tesla at speeds of 25 and 35 miles per hour (mph) versus a stationary pedestrian target in early, medium, and late FCW settings. Testing was also conducted with a 50% pedestrian offset and 75% offset conditions relative to the right side of the Tesla.

ISO 26262-1:2018 defines the fault tolerant time interval (FTTI) as the minimum time span from the occurrence of a fault within an electrical / electronic system to a possible occurrence of a hazardous event. FTTI provides a time limit within which compliant vehicle safety mechanisms must detect and react to faults capable of posing risk of harm to persons. This makes FTTI a vital safety characteristic for system design. Common automotive industry practice accommodates recording fault times of occurrence definitively. However, current practice for defining the time of hazardous event onset relies upon subjective judgements. This paper presents a novel method to define hazardous event onset more objectively. The method introduces the Streetscope Collision Hazard Measure (SHMTM) and a refined approach to hazardous event classification.

The design and location of rear-viewing mirrors or systems, and the presentation of the rear view to the driver can best be achieved if the designer and the engineer have adequate references available on the physiological functions of head and eye movements and on the perceptual capabilities of the human visual system. The following information and charts are provided for this purpose. For more complete information of the relationship of vision to forward vision, see SAE SP-279.

The improvement of vehicle soiling behavior has increasing interest over the past few years not only to satisfy customer requirements and ensure a good visibility of the surrounding traffic but also for autonomous vehicles, for which soiling investigation and improvement are even more important due to the demands of the cleanliness and induced functionality of the corresponding sensors. The main task is the improvement of the soiling behavior, i.e., reduction or even prevention of soiling of specific surfaces, for example, windows, mirrors, and sensors. This is mostly done in late stages of vehicle development and performed by experiments, e.g., wind tunnel tests, which are supplemented by simulation at an early development stage. Among other sources, the foreign soiling on the side mirror and the side window depend on the droplet detaching from the side mirror housing.

Abstract This article presents a novel approach to optimize the placement of light detection and ranging (LiDAR) sensors in autonomous driving vehicles using machine learning. As autonomous driving technology advances, LiDAR sensors play a crucial role in providing accurate collision data for environmental perception. The proposed method employs the deep deterministic policy gradient (DDPG) algorithm, which takes the vehicle’s surface geometry as input and generates optimized 3D sensor positions with predicted high visibility. Through extensive experiments on various vehicle shapes and a rectangular cuboid, the effectiveness and adaptability of the proposed method are demonstrated. Importantly, the trained network can efficiently evaluate new vehicle shapes without the need for re-optimization, representing a significant improvement over classical methods such as genetic algorithms.

The recent progress in camera-based technologies has prompted the development of prototype camera-based video systems, intended to replace conventional passenger vehicle mirrors. Given that a significant number of collisions during lane changes stem from drivers being unaware of nearby vehicles, these camera-based systems offer the potential to enhance safety. By affording drivers a broader field of view, they facilitate the detection of potential conflicts. This project was focused on analyzing naturalistic driving data in support of the Federal Motor Vehicle Safety Standard 111 regulatory endeavors. The goal was to assess the effectiveness and safety compatibility of prototype camera-based side-view systems as potential replacements for traditional side-view mirrors.

The windscreen wiping system is mandatory requirement for automotive vehicle as per Central motor vehicle rules (CMVR). The main scope of the standard is to ensure vision zones to be wiped by wiping system to ensure maximum field of vision to the driver. The evaluation of vision zones as per IS 15802:2008 is generally determined by virtual simulation by OEMs. The limitation of virtual simulation is due to actual tolerances in vehicle, due to seat fitment, ergonomic dimensions, seat cushioning effect and wiper non-effective operation which are not taken into consideration very well off. The testing methodology described in the paper is an in-house developed test method based on SAE recommended practices. With the help of 3D H-point machine and a laser based ‘Theodolite’ equipped with horizontal and vertical angle projections from single pivot point is used to develop various vision zones on an actual vehicle windscreen as per technical data.

Automated driving system is a multi-source sensor data fusion system. However different type sensor has different operating frequencies, different field of view, different detection capabilities and different sensor data transition delay. Aiming at these problems, this paper introduces the processing mechanism of out of sequence measurement data into the multi-target detection and tracking system based on millimeter wave radar and camera. After the comparison of ablation experiments, the longitudinal and lateral tracking performance of the fusion system is improved in different distance ranges.