This paper proposes a classification algorithm utilizing an open set recognition concept to conservatively detect lane change intention of surrounding vehicles. Conservatively predicting the lane change intention of the surrounding vehicles is needed to improve adaptive cruise control (ACC) performance and avoid possible accidents. However, existing machine learning can make incorrect decisions due to information not included in the training data set or confused data even with probability. To cope with this problem, we present a classification algorithm using a multi-class support vector machine applying an open set recognition concept to detect the surrounding vehicles' lane change intentions. Feature vectors are constructed from lateral information obtained by a Kalman filter using only radar and in-vehicle sensors. The open set recognition concept is adapted using Meta-Recognition based on binary classifiers scores. Furthermore, we analyze lateral information where an object vehicle changes lanes. From experimental results, we observe that the proposed system conservatively deals with wrong decisions and detects and cancels detecting the closest in-path vehicle (CIPV) earlier with average times of 1.4 sec and 0.4 sec compared with a commercial radar system, respectively.

The potential functions (PFs) have generally shown good performances in real-time path planning with computation efficiency conforming to the requirements of lower control systems in autonomous driving. However, several inherent limitations exist in using the PFs, including a local minimum in specific scenarios and no passage between closely spaced obstacles. Recent studies have focused on conventional scenarios where PFs are assumed to work normally, without malfunctioning, occurring during perilous situations. Therefore, we propose a specific safety tunnel (ST)-based model predictive controller (MPC) combined with PFs (PF-STMPC) to handle path-planning in extreme-emergency traffic scenarios (e.g., emergency braking and lane-changing obstacles). To further guarantee driving safety, we improve PFs with the responsibility-sensitive safety (RSS) model that accurately calculates the minimum safe longitudinal and lateral distances. Furthermore, a sigmoid-based ST is designed for emergency navigation if the PFs fail to plan a safe path due to the aforementioned inherent limitations, enabling the controller with planning functionality if necessary. The ST is embedded in the MPC-based tracking controller as a safe constraint sensitive to surrounding environments (e.g., road structure and obstacles). The proposed PF-STMPC was co-simulated using MATLAB/Simulink and CarSim Simulator under the constant speed condition. Compared with the state-of-the-art method, the proposed method demonstrated better performance in finding a safe path and eliminating severe yawing of the ego-vehicle (82.8% less in sideslip yawing amplitude and 57.7% shorter in the oscillation period of yaw angle) when facing traffic emergencies.

For accurate object vehicle estimation using radar, there are two fundamental problems: measurement uncertainties in calculating an object’s position with a virtual polygon box and latency due to commercial radar tracking algorithms. We present a data-driven object vehicle estimation scheme to solve measurement uncertainty and latency problems in radar systems. A radar accuracy model and latency coordination are proposed to reduce the tracking error. We first design data-driven radar accuracy models to improve the accuracy of estimation determined by the object vehicle’s position. The proposed model solves the measurement uncertainty problem within a feasible set for error covariance. The latency coordination is developed by analyzing the position error according to the relative velocity. The position error by latency is stored in a feasible set for relative velocity, and the solution is calculated from the given relative velocity. Removing the measurement uncertainty and latency of the radar system allows for a weighted interpolation to be applied to estimate the position of the object vehicle. Our method is tested by a scenario-based estimation experiment to validate the usefulness of the proposed data-driven object vehicle estimation scheme. We confirm that the proposed estimation method produces improved performance over the conventional radar estimation and previous methods.

This paper proposes a spatiotemporal architecture with a deep neural network (DNN) for road surface conditions and types classification using LiDAR for autonomous emergency braking (AEB) systems. It is known that LiDAR provides information on the reflectivity and number of point clouds depending on a road surface. Thus, this paper utilizes the characteristics of LiDAR and the vehicle’s speed obtained from in-vehicle sensors for classifying front road surface conditions for the AEB system application. First, we divided the front road area into four subregions to use spatial information from the vehicle’s speed. Second, we constructed feature vectors using each subregion’s reflectivity, number of point clouds, and in-vehicle information. Third, the DNN classifies road surface conditions and types for each subregion. Finally, the output of the DNN feeds into the spatiotemporal process to make the final classification reflecting vehicle speed and probability given by the outcomes of softmax functions of the DNN output layer. To validate the effectiveness of the proposed method, we performed a comparative study with five other algorithms. With the proposed DNN, we obtained the highest accuracy of 98.0% and 98.6% for two subregions near the vehicle. In addition, we implemented the proposed method on the NVIDIA Jetson TX2 board to confirm that it is applicable in real-time.

This paper presents a Spatio-Temporal Road Inference for a KEeping NETwork (StrikeNet), aimed at enhancing Road Lane Reconstruction (RLR) and lateral motion control in Autonomous Vehicles (AV) using deep neural networks. Accurate road lane model coefficients are essential for an effective Lane Keeping System (LKS), but the traditional vision system often fails in situations where lane markers are absent or faint and cannot be properly recognized. To overcome this, a driving dataset was restructured, combining road information from a vision system and forward images for spatial training of RLR. Sequential spatial learning outputs were then processed with in-vehicle sensor data for temporal inference via Long Short-Term Memory (LSTM). The StrikeNet was rigorously tested in both typical and uncertain driving environments. Comprehensive statistical and visualization analyses were conducted to evaluate the performance of various RLR methods and lateral motion control strategies. Remarkably, the RLR demonstrated its capability to derive reliable road coefficients even in the absence of lane markers. Upon performance comparison with four alternative techniques, our method yields the lowest error and variance between human steering inputs and the control input. Specifically, under high and low lane quality conditions, the proposed method maximally reduced the control input error by up to 72% and 66%, respectively, and decreased the variance by 54% and 94%, respectively. The findings highlight StrikeNet’s effectiveness in bolstering the fail-operational performance, and reliability of lane-keeping or lane departure warning systems in autonomous driving, thereby enhancing control continuity and mitigating path deviation-induced traffic accidents.

In this paper, a method for inferring the motion intentions of a neighboring vehicle ahead of an ego vehicle using a physics-informed deep neural network-based open-set classification approach is proposed. Relative motion data from real-world driving were categorized into known and unknown scenarios, with key feature trajectories represented as spatiotemporal 3D input data. A convolutional long short-term memory architecture was designed, and a novel loss function was proposed to incorporate physics-informed perspective and constraints, with integrated loss function’s convergence demonstrated for training. The proposed method was evaluated against comparative five classifiers in terms of 1) classification accuracy for known classes in single scenarios, 2) open-set classification accuracy, 3) analysis by deep reduced feature visualization, 4) generalization performance to unknown data, and 5) classification robustness and in-path decision validity in continuous scenarios. Results showed a 23.5% average improvement in accuracy, the highest generalization performance, and superior robustness, enabling faster and more reliable in-path detection compared to conventional radar including in ambiguous situations.