Abstracts

2020 IEEE Transactions on Intelligent Vehicles  
In this article we present a trajectory generation method for autonomous overtaking of unexpected obstacles in a dynamic urban environment. In these settings, blind spots can arise from perception limitations. For example when overtaking unexpected objects on the vehicle's ego lane on a two-way street. In this case, a human driver would first make sure that the opposite lane is free and that there is enough room to successfully execute the maneuver, and then it would cut into the opposite lane
more » ... n order to execute the maneuver successfully. We consider the practical problem of autonomous overtaking when the coverage of the perception system is impaired due to occlusion. Safe trajectories are generated by solving, in real-time, a non-linear constrained optimization, formulated as a receding horizon planner that maximizes the ego vehicle's visibility. The planner is complemented by a high-level behavior planner, which takes into account the occupancy of other traffic participants, the information from the vehicle's perception system, and the risk associated with the overtaking maneuver, to determine when the overtake maneuver should happen. The approach is validated in simulation and in experiments in real world traffic. In order to safely navigate highly dynamic scenarios, automated vehicles must be able to react quickly to changes in the environment and be able to understand trade-offs between lateral and longitudinal forces when limited by tire-road friction. We present a design and experimental validation of a nonlinear model predictive controller that is capable of handling these complex situations. By carefully selecting the vehicle model and mathematical encodings of the vehicle and obstacles, we enable the controller to quickly compute inputs while maintaining an accurate model of the vehicle's motion and its proximity to obstacles. Experimental results of a test vehicle performing an emergency double lane change to avoid two "pop-up" obstacles demonstrate the ability of the controller to coordinate lateral and longitudinal tire forces even in emergency situations when the tires are at their friction limits. The development of fully automating the operation of automobiles is rapidly progressing. However, challenges related to driving skill improvement of drivers exist because of the necessity of manual operation, and these are expected to continue in the future. Despite the fact that maintaining and improving manual driving performance will be indispensable in the near future, there are few studies on parking assistance systems aimed at improving the skills of the drivers. Therefore, we focused on steering timing as a candidate factor for skill improvement in reverse parking. In this article, we developed an assist method based on auditory presentation that indicates the timing for starting of steering in order to follow the target trajectory during reverse parking. From the results of verification tests of the assistance using the driving simulator, various effects such as improvement of driver's parking accuracy and reduction of workload were confirmed. Continuous estimation the driver's take-over readiness is critical for safe and timely transfer of control during the failure modes of autonomous vehicles. In this article, we propose a data-driven approach for estimating the driver's take-over readiness based purely on observable cues from in-vehicle vision sensors. We present an extensive naturalistic driving dataset of drivers in a conditionally autonomous vehicle running on Californian freeways. We collect subjective ratings for the driver's take-over readiness from multiple human observers viewing the sensor feed. Analysis of the ratings in terms of intra-class correlation coefficients (ICCs) shows a high degree of consistency in the ratings across raters. We define a metric for the driver's take-over readiness termed the 'Observable Readiness Index (ORI)' based on the ratings. Finally, we propose an LSTM model for continuous estimation of the driver's ORI based on a holistic representation of the driver's state, capturing gaze, hand, pose and foot activity. Our model estimates the ORI with a mean absolute error of 0.449 on a 5 point scale. Ride comfort and road holding are two substantial performance criteria related to the suspension system of road vehicles. These performance criteria are critical in controller design and chassis stability of automated driving vehicles. Control demands in terms of accuracy, quick response and robustness to matched and mismatched uncertainties are suggestive of employing adaptive robust controllers. Herein, a state observer-based modified sliding mode interval fuzzy type-2 neural network (FT2NN) controller is designed to suppress the vibrations from a typical rough terrain imposed to the nonlinear suspension system of the vehicles. The nonlinear system dynamics are estimated using the universal approximation capacity of the neuro-fuzzy type-2 approach and the states are obtained by the adaptive robust state observer. The membership functions (MFs) of the fuzzy type-2 system are employed to deal with the uncertainties through variable mean and variances for upper and lower MFs. Furthermore, the proposed controller has the advantage of relaxing the condition for the approximation error boundaries while the estimation step is utilized to observe the boundaries adaptively. A new adaptive compensator is employed to withstand the effect of the external disturbance, the approximation errors related to the unknown nonlinear functions and state estimations. The results obtained from the proposed controller are suggestive of the higher effectiveness of the proposed controller compared to the tested Neuro-PID controller, and also the passive suspension system. The high-fidelity MSC ADAMS based co-simulations were implemented to validate the practicality of the proposed controller. The lane-change strategy of autonomous vehicles is affected by its leading vehicle. It is necessary and challenging to simultaneously conduct the lane-change maneuvers while avoiding collisions with the leading vehicle. This article proposes a lane-change strategy considering the leading vehicle by synthesizing vehicle velocity prediction, motion planning, and trajectory tracking control. A scenario-based velocity prediction method using the input-output hidden Markov model (HMM) is proposed to predict the leading vehicle velocity. Then a motion planner integrating the predicted velocity is developed to generate the optimal trajectories for the lane-change maneuvers. The generated trajectories are tracked by a trajectory tracking controller. An improved composite nonlinear feedback (CNF) control algorithm is proposed to obtain smooth transient performances and fast responses. Human driver tests on a driving simulator show the leading vehicle velocity can be predicted by the proposed method. The motion planner and the trajectory tracking controller are validated in the CarSim simulations. The collision-free optimal trajectories are generated and tracked by the motion planner and the improved CNF controller. A systematic lane-change strategy considering the leading vehicle is effectively developed in this study.
doi:10.1109/tiv.2020.2971873 fatcat:af7apwjygrajxeenbiwv7grzii