Framework For Autonomous Driving Research Articles

FRENETIX: A High-Performance and Modular Motion Planning Framework for Autonomous Driving

FRENETIX: A High-Performance and Modular Motion Planning Framework for Autonomous Driving

A Declarative Metamorphic Testing Framework for Autonomous Driving

Autonomous driving has gained much attention from both industry and academia. Currently, Deep Neural Networks (DNNs) are widely used for perception and control in autonomous driving. However, several fatal accidents caused by autonomous vehicles have raised serious safety concerns about autonomous driving models. Some recent studies have successfully used the metamorphic testing technique to detect thousands of potential issues in some popularly used autonomous driving models. However, prior study is limited to a small set of metamorphic relations, which do not reflect rich, real-world traffic scenarios and are also not customizable. This paper presents a novel declarative rule-based metamorphic testing framework called RMT. RMT provides a rule template with natural language syntax, allowing users to flexibly specify an enriched set of testing scenarios based on real-world traffic rules and domain knowledge. RMT automatically parses human-written rules to metamorphic relations using an NLP-based rule parser referring to an ontology list and generates test cases with a variety of image transformation engines. We evaluated RMT on three autonomous driving models. With an enriched set of metamorphic relations, RMT detected a significant number of abnormal model predictions that were not detected by prior work. Through a large-scale human study on Amazon Mechanical Turk, we further confirmed the authenticity of test cases generated by RMT and the validity of detected abnormal model predictions.

Intention Prediction and Mixed Strategy Nash Equilibrium-Based Decision-Making Framework for Autonomous Driving in Uncontrolled Intersection

Decision-making in uncontrolled intersection is one of the main challenges in urban autonomous driving. This paper proposed a new decision-making framework in uncontrolled intersection based on the intention prediction method and Mixed Strategy Nash Equilibrium theory. The framework is a three-stage method: target vehicle motion prediction, driving mode decision, and motion planning. The driving intention (left turn, right turn, or go straight) of the target vehicle at the intersection can be predicted using the combination algorithm of GMM-HMM and SVM. According to the driving intention and road structure, the trajectory fitting module would use the Bezier curve to fit the predicted trajectory of a target vehicle. Furthermore, combined with trajectories of the ego vehicle, the S-T diagram is used to judge whether there is a spatio-temporal conflict point of the target vehicle and ego vehicle. If there is a conflict point of the target vehicle and ego vehicle, the driving mode (‘yield’ or ‘cross’) of the ego vehicle is selected by using the Mixed Strategy Nash Equilibrium theory. This method can not only avoid premature or unnecessary deceleration due to conservation but also avoid a collision or violent deceleration due to greed. According to the driving mode, the planning module uses the model predictive control algorithm to determine the optimal acceleration strategy. Have been verified by the vehicle test, it indicates that the proposed decision-making framework can make ego vehicles pass through the intersection safely and comfortably.

A Hierarchical Motion Planning Framework for Autonomous Driving in Structured Highway Environments

This paper presents an efficient hierarchical motion planning framework with a long planning horizon for autonomous driving in structured environments. A 3D motion planning with time information is a non-convex problem because there exists more than one local minimum point and various constraints such as roads and obstacles. Accordingly, to deal with high computational complexity and the problem of falling into a local minimum in an enormous solution space, a decoupled method is utilized, that consists of two steps: <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Long-term</i> planning and <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">short-term</i> planning. First, the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">long-term</i> planner provides reasonable far-sighted behavior through two processes. In the first process, a rough path that includes a driving strategy is generated in the 2D spatial space. Then, the jump point search algorithm is utilized with time information on the path to reduce the computational burden of A*, giving an acceptable quality of solution at the same time. In this step, a safe, comfortable, and dynamically feasible trajectory is generated. Next, the <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">short-term</i> planner optimizes a short-sighted trajectory using particle swarm optimization. In this method, a steering angle set is encoded as a particle, resulting in a safe, comfortable, and kinodynamically feasible trajectory. The proposed algorithm is implemented and evaluated in a range of vehicle-in-the-loop simulation scenarios, which include various virtual static and dynamic obstacles generated by Intelligent Driver Model. In the evaluation results, the proposed method reduced the computation time by up to 0.696 s with increasing the step cost by up to about 3%. The proposed algorithm is executed every 100 ms for a planning horizon of 10 seconds, and the average computation time is 31 ms with the worst-case computation time of 94 ms.

An Optimized Trajectory Planner and Motion Controller Framework for Autonomous Driving in Unstructured Environments

This paper proposes an optimized trajectory planner and motion planner framework, which aim to deal with obstacle avoidance along a reference road for autonomous driving in unstructured environments. The trajectory planning problem is decomposed into lateral and longitudinal planning sub-tasks along the reference road. First, a vehicle kinematic model with road coordinates is established to describe the lateral movement of the vehicle. Then, nonlinear optimization based on a vehicle kinematic model in the space domain is employed to smooth the reference road. Second, a multilayered search algorithm is applied in the lateral-space domain to deal with obstacles and find a suitable path boundary. Then, the optimized path planner calculates the optimal path by considering the distance to the reference road and the curvature constraints. Furthermore, the optimized speed planner takes into account the speed boundary in the space domain and the constraints on vehicle acceleration. The optimal speed profile is obtained by using a numerical optimization method. Furthermore, a motion controller based on a kinematic error model is proposed to follow the desired trajectory. Finally, the experimental results show the effectiveness of the proposed trajectory planner and motion controller framework in handling typical scenarios and avoiding obstacles safely and smoothly on the reference road and in unstructured environments.

Target Vehicle Motion Prediction-Based Motion Planning Framework for Autonomous Driving in Uncontrolled Intersections

This paper presents a motion-planning framework for urban autonomous driving at uncontrolled intersections. The intention and future state of the target vehicles are predicted using information obtained from the environment sensors. The target state prediction module employs an Interacting Multiple Model (IMM) filter to infer the intention of targets. The prediction results of each model of the IMM filter are fused to predict the future state of targets. The proposed predictor uses the intelligent driver model based-driver behavior model to construct the local filter of IMM. The driving mode decision is realized as a state machine consisting of two phases, ‘Approach’ and ‘Risk Management’. The risk management phase is composed of two sub-modes, ‘Cross’ and ‘Yield’. The state transition conditions between phases and modes are defined by introducing the concepts of ‘Critical gap’ and ‘Follow-up gap’. Based on the determined driving mode, the motion planning module consists of two sub-modules for each phase. The required deceleration determination for the approach phase is proposed to consider the occluded region in order to prevent inevitable collisions caused by fast approaches. The model predictive controller for the risk management phase is designed to determine the desired acceleration to guarantee safety and prevent unnecessary deceleration simultaneously. Both computer simulation studies and vehicle tests are conducted to evaluate the proposed framework. The results indicate that the proposed framework ensures the safety at uncontrolled intersections with a driving pattern similar to that of a driver.

Clothoid: An Integrated Hierarchical Framework for Autonomous Driving in a Dynamic Urban Environment.

In recent years, research and development of autonomous driving technology have gained much interest. Many autonomous driving frameworks have been developed in the past. However, building a safely operating fully functional autonomous driving framework is still a challenge. Several accidents have been occurred with autonomous vehicles, including Tesla and Volvo XC90, resulting in serious personal injuries and death. One of the major reasons is the increase in urbanization and mobility demands. The autonomous vehicle is expected to increase road safety while reducing road accidents that occur due to human errors. The accurate sensing of the environment and safe driving under various scenarios must be ensured to achieve the highest level of autonomy. This research presents Clothoid, a unified framework for fully autonomous vehicles, that integrates the modules of HD mapping, localization, environmental perception, path planning, and control while considering the safety, comfort, and scalability in the real traffic environment. The proposed framework enables obstacle avoidance, pedestrian safety, object detection, road blockage avoidance, path planning for single-lane and multi-lane routes, and safe driving of vehicles throughout the journey. The performance of each module has been validated in K-City under multiple scenarios where Clothoid has been driven safely from the starting point to the goal point. The vehicle was one of the top five to successfully finish the autonomous vehicle challenge (AVC) in the Hyundai AVC.

VRDriving: A Virtual-to-Real Autonomous Driving Framework Based on Adversarial Learning

Due to the high cost of driving testing and the data collection for autonomous cars in the real world, transferring driving models from the simulator to the real world is significant and urgent. This article proposes a novel virtual-to-real driving (VRDriving) framework based on adversarial learning, which is able to transfer the ability of end-to-end learning models trained by simulation virtual data sets to the real world. The framework consists of three parts: 1) an end-to-end learning-based estimator with the cost-sensitive loss function for driving commands estimation; 2) an encoder for extracting the driving domain features from unlabeled real driving images; and 3) a discriminator for distinguishing whether the virtual driving domain features are real or fake. The above three modules can be trained simultaneously. With such a novel learning framework, the driving domain features encoded from the real images are closer to the driving domain features inside the end-to-end estimator via adversarial learning mechanism. Three experiments show the proposed VRDriving framework performs the state-of-the-art estimation accuracy of the steering command than several baselines and other typical frameworks. Therefore, the VRDriving framework is promising to improve the development efficiency of future autonomous driving.

Decision-Making Framework for Autonomous Driving at Road Intersections: Safeguarding Against Collision, Overly Conservative Behavior, and Violation Vehicles

In this paper, we propose a decision-making framework for autonomous driving at road intersections that determines appropriate maneuvers for an autonomous vehicle to navigate an intersection safely and efficiently (regarding making progress ), even in the face of violation vehicles—one of the most challenging tasks in the domain of autonomous vehicles. The proposed framework uses a digital map to predict future paths of observed vehicles and then uses the predicted future paths to identify potential threats (vehicles) and collision areas, regardless of whether observed vehicles are obeying traffic rules at the intersection. Next, under an independent and distributed reasoning structure, it systematically, reliably, and robustly assesses the potential threats, even under incomplete and uncertain noise data, by way of a threat measure, Bayesian networks, and time window filtering. It then uses this information to determine appropriate maneuvers for the autonomous vehicle to navigate the intersection safely and efficiently. We have tested and evaluated the proposed framework through in-vehicle testing on a closed urban test road under traffic conditions inclusive of nonviolation and violation vehicles. In-vehicle testing results show the performance of the proposed framework to be sufficiently reliable for autonomous driving at intersections regarding reliability, robustness, safety, and efficiency.

Deep Reinforcement Learning framework for Autonomous Driving

Reinforcement learning is considered to be a strong AI paradigm which can be used to teach machines through interaction with the environment and learning from their mistakes. Despite its perceived utility, it has not yet been successfully applied in automotive applications. Motivated by the successful demonstrations of learning of Atari games and Go by Google DeepMind, we propose a framework for autonomous driving using deep reinforcement learning. This is of particular relevance as it is difficult to pose autonomous driving as a supervised learning problem due to strong interactions with the environment including other vehicles, pedestrians and roadworks. As it is a relatively new area of research for autonomous driving, we provide a short overview of deep reinforcement learning and then describe our proposed framework. It incorporates Recurrent Neural Networks for information integration, enabling the car to handle partially observable scenarios. It also integrates the recent work on attention models to focus on relevant information, thereby reducing the computational complexity for deployment on embedded hardware. The framework was tested in an open source 3D car racing simulator called TORCS. Our simulation results demonstrate learning of autonomous maneuvering in a scenario of complex road curvatures and simple interaction of other vehicles.