scholarly journals Autonomous Driving in Traffic: Boss and the Urban Challenge

AI Magazine ◽  
2009 ◽  
Vol 30 (2) ◽  
pp. 17 ◽  
Author(s):  
Chris Urmson ◽  
Chris Baker ◽  
John Dolan ◽  
Paul Rybski ◽  
Bryan Salesky ◽  
...  
Keyword(s):  
Real Time ◽  
Autonomous Vehicles ◽  
Autonomous Vehicle ◽  
Autonomous Driving ◽  
Software System ◽  
Intelligent Software ◽  
The World

The DARPA Urban Challenge was a competition to develop autonomous vehicles capable of safely, reliably and robustly driving in traffic. In this article we introduce Boss, the autonomous vehicle that won the challenge. Boss is complex artificially intelligent software system embodied in a 2007 Chevy Tahoe. To navigate safely, the vehicle builds a model of the world around it in real time. This model is used to generate safe routes and motion plans in both on roads and in unstructured zones. An essential part of Boss’ success stems from its ability to safely handle both abnormal situations and system glitches.

scholarly journals On the Performance of One-Stage and Two-Stage Object Detectors in Autonomous Vehicles Using Camera Data

2020 ◽  
Vol 13 (1) ◽  
pp. 89
Author(s):  
Manuel Carranza-García ◽  
Jesús Torres-Mateo ◽  
Pedro Lara-Benítez ◽  
Jorge García-Gutiérrez
Keyword(s):  
Deep Learning ◽  
Real Time ◽  
Autonomous Vehicles ◽  
Remote Sensing Data ◽  
Autonomous Driving ◽  
Two Stage ◽  
Detection Systems ◽  
One Stage ◽  
Time Requirements ◽  
Speed Accuracy

Object detection using remote sensing data is a key task of the perception systems of self-driving vehicles. While many generic deep learning architectures have been proposed for this problem, there is little guidance on their suitability when using them in a particular scenario such as autonomous driving. In this work, we aim to assess the performance of existing 2D detection systems on a multi-class problem (vehicles, pedestrians, and cyclists) with images obtained from the on-board camera sensors of a car. We evaluate several one-stage (RetinaNet, FCOS, and YOLOv3) and two-stage (Faster R-CNN) deep learning meta-architectures under different image resolutions and feature extractors (ResNet, ResNeXt, Res2Net, DarkNet, and MobileNet). These models are trained using transfer learning and compared in terms of both precision and efficiency, with special attention to the real-time requirements of this context. For the experimental study, we use the Waymo Open Dataset, which is the largest existing benchmark. Despite the rising popularity of one-stage detectors, our findings show that two-stage detectors still provide the most robust performance. Faster R-CNN models outperform one-stage detectors in accuracy, being also more reliable in the detection of minority classes. Faster R-CNN Res2Net-101 achieves the best speed/accuracy tradeoff but needs lower resolution images to reach real-time speed. Furthermore, the anchor-free FCOS detector is a slightly faster alternative to RetinaNet, with similar precision and lower memory usage.

scholarly journals Real - Time Trajectory and Velocity Planning for Autonomous Vehicles

2021 ◽  
Vol 10 (5) ◽  
pp. 439-448
Author(s):  
Hrishikesh Dey ◽  
Rithika Ranadive ◽  
Abhishek Chaudhari
Keyword(s):  
Path Planning ◽  
Velocity Profile ◽  
Real Time ◽  
Autonomous Vehicle ◽  
Optimal Path ◽  
Autonomous Driving ◽  
Navigation Algorithm ◽  
Velocity Planning ◽  
Planning Algorithm ◽  
Path Planner

Path planning algorithm integrated with a velocity profile generation-based navigation system is one of the most important aspects of an autonomous driving system. In this paper, a real-time path planning solution to obtain a feasible and collision-free trajectory is proposed for navigating an autonomous car on a virtual highway. This is achieved by designing the navigation algorithm to incorporate a path planner for finding the optimal path, and a velocity planning algorithm for ensuring a safe and comfortable motion along the obtained path. The navigation algorithm was validated on the Unity 3D Highway-Simulated Environment for practical driving while maintaining velocity and acceleration constraints. The autonomous vehicle drives at the maximum specified velocity until interrupted by vehicular traffic, whereas then, the path planner, based on the various constraints provided by the simulator using µWebSockets, decides to either decelerate the vehicle or shift to a more secure lane. Subsequently, a splinebased trajectory generation for this path results in continuous and smooth trajectories. The velocity planner employs an analytical method based on trapezoidal velocity profile to generate velocities for the vehicle traveling along the precomputed path. To provide smooth control, an s-like trapezoidal profile is considered that uses a cubic spline for generating velocities for the ramp-up and ramp-down portions of the curve. The acceleration and velocity constraints, which are derived from road limitations and physical systems, are explicitly considered. Depending upon these constraints and higher module requirements (e.g., maintaining velocity, and stopping), an appropriate segment of the velocity profile is deployed. The motion profiles for all the use-cases are generated and verified graphically.

scholarly journals Computer vision based obstacle detection and target tracking for autonomous vehicles

2021 ◽  
Vol 336 ◽  
pp. 07004
Author(s):  
Ruoyu Fang ◽  
Cheng Cai
Keyword(s):  
Neural Network ◽  
Computer Vision ◽  
Deep Learning ◽  
Target Tracking ◽  
Real Time ◽  
Autonomous Vehicles ◽  
Autonomous Vehicle ◽  
Obstacle Detection ◽  
Pid Algorithm ◽  
Deep Learning Neural Network

Obstacle detection and target tracking are two major issues for intelligent autonomous vehicles. This paper proposes a new scheme to achieve target tracking and real-time obstacle detection of obstacles based on computer vision. ResNet-18 deep learning neural network is utilized for obstacle detection and Yolo-v3 deep learning neural network is employed for real-time target tracking. These two trained models can be deployed on an autonomous vehicle equipped with an NVIDIA Jetson Nano motherboard. The autonomous vehicle moves to avoid obstacles and follow tracked targets by camera. Adjusting the steering and movement of the autonomous vehicle according to the PID algorithm during the movement, therefore, will help the proposed vehicle achieve stable and precise tracking.

scholarly journals On the Development of Autonomous Vehicle Safety Distance by an RSS Model Based on a Variable Focus Function Camera

Sensors ◽  
2021 ◽  
Vol 21 (20) ◽  
pp. 6733
Author(s):  
Min-Joong Kim ◽  
Sung-Hun Yu ◽  
Tong-Hyun Kim ◽  
Joo-Uk Kim ◽  
Young-Min Kim
Keyword(s):  
Autonomous Vehicles ◽  
Focal Length ◽  
Autonomous Vehicle ◽  
Autonomous Driving ◽  
Cruise Control ◽  
Radar Sensors ◽  
Safety And Reliability ◽  
Time Required ◽  
Variable Focus ◽  
Camera Sensors

Today, a lot of research on autonomous driving technology is being conducted, and various vehicles with autonomous driving functions, such as ACC (adaptive cruise control) are being released. The autonomous vehicle recognizes obstacles ahead by the fusion of data from various sensors, such as lidar and radar sensors, including camera sensors. As the number of vehicles equipped with such autonomous driving functions increases, securing safety and reliability is a big issue. Recently, Mobileye proposed the RSS (responsibility-sensitive safety) model, which is a white box mathematical model, to secure the safety of autonomous vehicles and clarify responsibility in the case of an accident. In this paper, a method of applying the RSS model to a variable focus function camera that can cover the recognition range of a lidar sensor and a radar sensor with a single camera sensor is considered. The variables of the RSS model suitable for the variable focus function camera were defined, the variable values were determined, and the safe distances for each velocity were derived by applying the determined variable values. In addition, as a result of considering the time required to obtain the data, and the time required to change the focal length of the camera, it was confirmed that the response time obtained using the derived safe distance was a valid result.

scholarly journals Automated Testing of Ultrawideband Positioning for Autonomous Driving

2020 ◽  
Vol 2020 ◽  
pp. 1-15
Author(s):  
Benjamin Vedder ◽  
Bo Joel Svensson ◽  
Jonny Vinter ◽  
Magnus Jonsson
Keyword(s):  
Fault Tolerance ◽  
Real Time ◽  
Autonomous Vehicles ◽  
Fault Injection ◽  
Autonomous Driving ◽  
Automated Testing ◽  
Test Cases ◽  
Positioning System ◽  
Novel Approach ◽  
Driving Model

Autonomous vehicles need accurate and dependable positioning, and these systems need to be tested extensively. We have evaluated positioning based on ultrawideband (UWB) ranging with our self-driving model car using a highly automated approach. Random drivable trajectories were generated, while the UWB position was compared against the Real-Time Kinematic Satellite Navigation (RTK-SN) positioning system which our model car also is equipped with. Fault injection was used to study the fault tolerance of the UWB positioning system. Addressed challenges are automatically generating test cases for real-time hardware, restoring the state between tests, and maintaining safety by preventing collisions. We were able to automatically generate and carry out hundreds of experiments on the model car in real time and rerun them consistently with and without fault injection enabled. Thereby, we demonstrate one novel approach to perform automated testing on complex real-time hardware.

scholarly journals SAGE: A S plit- A rchitecture Methodolo g y for E fficient End-to-End Autonomous Vehicle Control

2021 ◽  
Vol 20 (5s) ◽  
pp. 1-22
Author(s):  
Arnav Malawade ◽  
Mohanad Odema ◽  
Sebastien Lajeunesse-degroot ◽  
Mohammad Abdullah Al Faruque
Keyword(s):  
High Resolution ◽  
Real Time ◽  
Autonomous Vehicles ◽  
Autonomous Vehicle ◽  
Industry Standard ◽  
Wide Range ◽  
Energy Consuming ◽  
Time Latency ◽  
4G Lte ◽  
Hardware Platforms

Autonomous vehicles (AV) are expected to revolutionize transportation and improve road safety significantly. However, these benefits do not come without cost; AVs require large Deep-Learning (DL) models and powerful hardware platforms to operate reliably in real-time, requiring between several hundred watts to one kilowatt of power. This power consumption can dramatically reduce vehicles’ driving range and affect emissions. To address this problem, we propose SAGE: a methodology for selectively offloading the key energy-consuming modules of DL architectures to the cloud to optimize edge, energy usage while meeting real-time latency constraints. Furthermore, we leverage Head Network Distillation (HND) to introduce efficient bottlenecks within the DL architecture in order to minimize the network overhead costs of offloading with almost no degradation in the model’s performance. We evaluate SAGE using an Nvidia Jetson TX2 and an industry-standard Nvidia Drive PX2 as the AV edge, devices and demonstrate that our offloading strategy is practical for a wide range of DL models and internet connection bandwidths on 3G, 4G LTE, and WiFi technologies. Compared to edge-only computation, SAGE reduces energy consumption by an average of 36.13% , 47.07% , and 55.66% for an AV with one low-resolution camera, one high-resolution camera, and three high-resolution cameras, respectively. SAGE also reduces upload data size by up to 98.40% compared to direct camera offloading.

scholarly journals Autonomous Driving—A Crash Explained in Detail

2019 ◽  
Vol 9 (23) ◽  
pp. 5126 ◽  
Author(s):  
Betz ◽  
Heilmeier ◽  
Wischnewski ◽  
Stahl ◽  
Lienkamp
Keyword(s):  
Autonomous Vehicles ◽  
Autonomous Vehicle ◽  
Autonomous Driving ◽  
Software Test ◽  
Planning And Control ◽  
Modular Software ◽  
Fatal Crash ◽  
Detailed Explanation ◽  
A Minor ◽  
And Control

Since 2017, a research team from the Technical University of Munich has developed a software stack for autonomous driving. The software was used to participate in the Roborace Season Alpha Championship. The championship aims to achieve autonomous race cars competing with different software stacks against each other. In May 2019, during a software test in Modena, Italy, the greatest danger in autonomous driving became reality: A minor change in environmental influences led an extensively tested software to crash into a barrier at speed. Crashes with autonomous vehicles have happened before but a detailed explanation of why software failed and what part of the software was not working correctly is missing in research articles. In this paper we present a general method that can be used to display an autonomous vehicle disengagement to explain in detail what happened. This method is then used to display and explain the crash from Modena. Firstly a brief introduction into the modular software stack that was used in the Modena event, consisting of three individual parts—perception, planning, and control—is given. Furthermore, the circumstancescausing the crash are elaborated in detail. By presented and explaining in detail which softwarepart failed and contributed to the crash we can discuss further software improvements. As a result, we present necessary functions that need to be integrated in an autonomous driving software stack to prevent such a vehicle behavior causing a fatal crash. In addition we suggest an enhancement of the current disengagement reports for autonomous driving regarding a detailed explanation of the software part that was causing the disengagement. In the outlook of this paper we present two additional software functions for assessing the tire and control performance of the vehicle to enhance the autonomous.

From Neurorobotic Localization to Autonomous Vehicles

2019 ◽  
Vol 07 (03) ◽  
pp. 183-194
Author(s):  
Yoan Espada ◽  
Nicolas Cuperlier ◽  
Guillaume Bresson ◽  
Olivier Romain
Keyword(s):  
Visual Information ◽  
Autonomous Vehicles ◽  
Autonomous Vehicle ◽  
Real Data ◽  
Autonomous Driving ◽  
Recognition System ◽  
Indoor Navigation ◽  
Learning System ◽  
Place Recognition ◽  
Outdoor Environments

The navigation of autonomous vehicles is confronted to the problem of an efficient place recognition system which is able to handle outdoor environments on the long run. The current Simultaneous Localization and Mapping (SLAM) and place recognition solutions have limitations that prevent them from achieving the performances needed for autonomous driving. This paper suggests handling the problem from another perspective by taking inspiration from biological models. We propose a neural architecture for the localization of an autonomous vehicle based on a neurorobotic model of the place cells (PC) found in the hippocampus of mammals. This model is based on an attentional mechanism and only takes into account visual information from a mono-camera and the orientation information to self-localize. It has the advantage to work with low resolution camera without the need of calibration. It also does not need a long learning phase as it uses a one-shot learning system. Such a localization model has already been integrated in a robot control architecture which allows for successful navigation both in indoor and small outdoor environments. The contribution of this paper is to study how it passes the scale change by evaluating the performance of this model over much larger outdoor environments. Eight experiments using real data (image and orientation) grabbed by a moving vehicle are studied (coming from the KITTI odometry datasets and datasets taken with VEDECOM vehicles). Results show the strong adaptability to different kinds of environments of this bio-inspired model primarily developed for indoor navigation.

scholarly journals Visual Semantic Landmark-Based Robust Mapping and Localization for Autonomous Indoor Parking

Sensors ◽  
2019 ◽  
Vol 19 (1) ◽  
pp. 161 ◽  
Author(s):  
Junqiao Zhao ◽  
Yewei Huang ◽  
Xudong He ◽  
Shaoming Zhang ◽  
Chen Ye ◽  
...  
Keyword(s):  
Real Time ◽  
Autonomous Vehicles ◽  
Indoor Localization ◽  
Autonomous Driving ◽  
Human Beings ◽  
Low Level ◽  
Localization And Mapping ◽  
Challenging Tasks ◽  
Parking Lot ◽  
Feature Based

Autonomous parking in an indoor parking lot without human intervention is one of the most demanded and challenging tasks of autonomous driving systems. The key to this task is precise real-time indoor localization. However, state-of-the-art low-level visual feature-based simultaneous localization and mapping systems (VSLAM) suffer in monotonous or texture-less scenes and under poor illumination or dynamic conditions. Additionally, low-level feature-based mapping results are hard for human beings to use directly. In this paper, we propose a semantic landmark-based robust VSLAM for real-time localization of autonomous vehicles in indoor parking lots. The parking slots are extracted as meaningful landmarks and enriched with confidence levels. We then propose a robust optimization framework to solve the aliasing problem of semantic landmarks by dynamically eliminating suboptimal constraints in the pose graph and correcting erroneous parking slots associations. As a result, a semantic map of the parking lot, which can be used by both autonomous driving systems and human beings, is established automatically and robustly. We evaluated the real-time localization performance using multiple autonomous vehicles, and an repeatability of 0.3 m track tracing was achieved at a 10 kph of autonomous driving.

scholarly journals Processor-in-the-Loop Architecture Design and Experimental Validation for an Autonomous Racing Vehicle

2021 ◽  
Vol 11 (16) ◽  
pp. 7225
Author(s):  
Eugenio Tramacere ◽  
Sara Luciani ◽  
Stefano Feraco ◽  
Angelo Bonfitto ◽  
Nicola Amati
Keyword(s):  
Real Time ◽  
Personal Computer ◽  
Trajectory Planning ◽  
Autonomous Vehicles ◽  
High Performance ◽  
Experimental Validation ◽  
Low Cost ◽  
Autonomous Driving ◽  
Obstacle Detection ◽  
System Level

Self-driving vehicles have experienced an increase in research interest in the last decades. Nevertheless, fully autonomous vehicles are still far from being a common means of transport. This paper presents the design and experimental validation of a processor-in-the-loop (PIL) architecture for an autonomous sports car. The considered vehicle is an all-wheel drive full-electric single-seater prototype. The retained PIL architecture includes all the modules required for autonomous driving at system level: environment perception, trajectory planning, and control. Specifically, the perception pipeline exploits obstacle detection algorithms based on Artificial Intelligence (AI), and the trajectory planning is based on a modified Rapidly-exploring Random Tree (RRT) algorithm based on Dubins curves, while the vehicle is controlled via a Model Predictive Control (MPC) strategy. The considered PIL layout is implemented firstly using a low-cost card-sized computer for fast code verification purposes. Furthermore, the proposed PIL architecture is compared in terms of performance to an alternative PIL using high-performance real-time target computing machine. Both PIL architectures exploit User Datagram Protocol (UDP) protocol to properly communicate with a personal computer. The latter PIL architecture is validated in real-time using experimental data. Moreover, they are also validated with respect to the general autonomous pipeline that runs in parallel on the personal computer during numerical simulation.

Sign in / Sign up

Export Citation Format

Share Document