Compared to conventional 3D radar, the 4D imaging radar provides additional height data and finer resolution measurements. Moreover, compared to LiDAR sensors, 4D imaging radar is more cost-effective and offers enhanced durability against challenging weather conditions. Despite these advantages, radar-based localization systems face several challenges, including limited resolution, leading to scattered object recognition and less precise localization. Additionally, existing methods that form submaps from filtered results can accumulate errors, leading to blurred submaps and reducing the accuracy of the SLAM and odometry. To address these challenges, this paper introduces Radar4VoxMap, a novel approach designed to enhance radar-only odometry. The method includes an RCS-weighted voxel distribution map that improves registration accuracy. Furthermore, fixed-lag optimization with the graph is used to optimize both the submap and pose, effectively reducing cumulative errors. The proposed method has shown strong performance on open datasets. The code is available at: https://github.com/ailab-hanyang/Radar4VoxMap

The importance of precise localization technology for the autonomous driving of industrial mobile robots is steadily increasing. Notably, research into enhancing accuracy and robustness by fusing multiple systems is actively conducted rather than relying on a single localization system. We highlight the use of track-to-track (T2T) fusion, which takes the localization results of independent systems as input. This approach eliminates system adjustments with sensor changes, offering benefits for industrial mobile robots. However, existing T2T-based fusion methods suffer from overlooking slowly changing biases that can gradually increase over time due to sensor drift errors, map biases, etc. Since biases have different values and frequencies for each system, they are challenging for conventional T2T methods to handle. This article proposes a localization fusion framework that tackles such slowly varying biases. First, estimating the distinct biases inherent to each system poses a challenging problem; therefore, we align them to a single common bias. Second, localization estimates with a common bias are fused using a split covariance intersection filter, one of the T2T fusion techniques, considering the independence and correlation within each system to ensure fusion consistency. The proposed method has been validated in both simulation and real-world environments, confirming superior performance compared to existing algorithms.

The development of autonomous vehicles has been accelerating, marked by a variety of competitions that challenge teams with diverse missions. Recently, racing-based autonomous driving competitions have gained prominence. Notably, the 2023 Hyundai Motor Group Autonomous Driving Challenge (HMG ADC) stands out as a manufacturer-operated event with a racing concept. This competition was distinctive, featuring mass-produced vehicles on race track with multiple vehicles simultaneously. In this paper, we explore the AutoKU team’s participation in the HMG ADC, highlighting their system, which is designed for two types of driving: solo and multi-vehicle racing. We detail the use of an identical mass-produced Hyundai IONIQ 5 vehicle equipped for autonomous driving without any performance modifications. The paper will discuss AutoKU’s approach and performance in solo and multi-vehicle races, showcasing their strategies and achievements in this innovative autonomous racing challenge. (Video: https://youtu.be/wLtmUkahnYA?si=AjqH6hYe10O94laq).

A computing system for autonomous vehicles must efficiently process vast amounts of data from various sensors in real-time and have a backup design to handle system failures. Adding more computing devices improves performance and redundancy, but increases costs and energy consumption. With the development of vehicle-to-everything (V2X) communication technologies and edge computing, such as multiaccess edge computing (MEC), computation offloading has been introduced. This process involves the use of cloud or edge servers to handle calculations normally performed by on-board computers, thereby reducing their workload and offering redundancy. This article proposes an offloading strategy that optimizes the allocation of software components (SWCs) in autonomous driving software. By optimizing SWC allocation, SWCs can be effectively assigned to specific computing units. This article established a cost function and constraints based on SWC-specific features to address the offloading decision problem as an allocation optimization issue. This article also defines safety-related metrics, including the response time requirements and failure risks of SWCs, to set criteria for offloading decisions. The proposed SWC optimization method is tested and validated using a real autonomous driving vehicle demonstration involving both cloud and vehicular edge environments.

In the realm of autonomous driving, motion planning for the ego vehicle necessitates the prediction of surrounding vehicles’ motions. This prediction traditionally relies on object tracking modules containing several sensors to gauge vehicle positions and velocities. However, existing physical or maneuver-based model approaches overlook important aspects of vehicle interactions that significantly affect actual vehicle movement. Ignoring these interactions can lead to inadequate ego vehicle’s motion planning. Addressing this gap, this paper proposes a novel approach: the Collision Probability Field (CPF)-based interaction-aware longitudinal motion prediction. Our methodology uniquely integrates the CPF, derived from the uncertainty of sensing information, to account for the probabilistic state of vehicle positions and velocities. This allows the prediction algorithm to consider not just the static data, but also the dynamic interactions between vehicles such as collision. Our approach was tested in various scenarios, including lane changes with an approaching vehicle from behind and different driver behavior models in real-world conditions. Our findings demonstrate a significant improvement in prediction accuracy for the motion planning of ego vehicle, highlighting the importance of interaction-aware predictions in autonomous driving systems.

The localization system is an essential element in robotics, which can provide accurate position information. Multiple localization systems can be integrated for reliable localization operations because there are various methods for measuring the position or processing algorithms. Significantly, the track-to-track (T2T) fusion method can fuse multiple localization systems using each system's estimate without accessing the sensor's low data. However, most T2T fusion-based localization systems ignore slowly varying biases, such as drift errors, odometry errors, and offsets among multiple maps. This can degrade the localization performance because a slowly varying bias is directly reflected in the localization estimate. Therefore, a slowly varying bias must be considered in the fusion process to derive reliable estimates. This study proposes a T2T fusion-based localization system that considers a slowly varying bias. First, the slow-varying bias difference between the systems was estimated. Because each localization system can have a different bias, the estimated bias difference was used to align it with the reference system. Second, a fused estimate can be obtained by T2T fusion using biasaligned estimates. The proposed fusion system can also be used without limiting the number of inputs to the localization system. The proposed system was compared with various T2T-based localization fusion algorithms for verification in a simulation environment, and it exhibited the best performance in RMSE error comparison.

This paper presents a hierarchical framework combining a machine learning (ML)-based approach with a model-based approach to predict the behavior and trajectory of surrounding target vehicles on a highway. First, the behavior predictor based on a recurrent neural network determines the behavior of the target vehicle by learning its complex interactions with surrounding vehicles and the traffic environment. Then, the trajectory predictor generates a predicted trajectory which follows the predicted behavior for each target vehicle. A curvature continuous spiral curve and model predicted control are used for the trajectory predictor to consider the dynamic constraints and the collision safety of the target vehicle. The hierarchical predictor composed of the ML-based approach and the model-based approach can predict the behavior and trajectory of a target vehicle, taking into account dynamic constraints and collisions as well as complex interactions with surrounding traffic. We evaluated the proposed predictor through NGSIM public dataset. The results showed that the predicted trajectories have lower errors over a long prediction time. We also showed that the proposed predictor could operate in real-time by efficiently utilizes computing resources.

Precise localization in map-changing environments is a major challenge faced in the case of autonomous vehicles. In such environments, map-matching-based localization can result in an incorrect vehicle pose because the High-Definition map (HD map) and the sensor measurements from the real environments are different. In order to solve this problem, this paper proposes robust localization in map-changing environments based on a hierarchical approach of sliding window optimization and filtering. The changing environments are explicitly modeled by a sub-map from the sliding-window-based graph optimization, and the generated sub-map is used for the map matching of the real-time filter. Since the optimized sub-map includes sensor measurement from both the changed and unchanged regions, it reduces the proportion of the changed region in the total matching region, thereby increasing the robustness of the map matching. The proposed hierarchical localization algorithm is verified and evaluated via simulation and experiments. The results show that the proposed algorithm provides a robust vehicle pose in map-changing environments.

A safe trajectory planning for on-road autonomous driving is a challenging problem owing to the variety and complexity of driving environments. The problem should involve the consideration of numerous aspects such as road geometry, lane-structured roads, traffic regulations, traffic participants, and vehicle physical limitations. Furthermore, dynamically changing movements of surrounding vehicles make the problem more challenging. It requires the planner's ability to react to the changes in the driving environments in real time. To solve this problem, sampling and numerical optimization-based trajectory planners were introduced. However, these methods have their own limitations in generating a safe trajectory in these dynamic scenarios. To overcome these issues, this paper proposes a hybrid trajectory planning scheme to integrate the strength of the sampling and optimization methods. With the sampling method for a lateral movement, the planner can deal with various trajectories with multiple maneuvers. This helps the planner to generate a reactive trajectory in a dynamically changing environment. The numerical optimization of a longitudinal movement enables the planner to adapt to diverse situations without restriction of predefined patterns for specific driving purposes. The proposed method was implemented with an embedded optimization coder and C++ environment. Based on this, its performance was evaluated through simulation and real driving tests in various on-road dynamic scenarios.

This paper presents track fusion and behavioral reasoning for moving vehicles in close proximity based on the curvilinear coordinates of roadway geometries. The inferred track and behavior of other vehicles can be used to perform safe actions in intelligent vehicle applications and autonomous driving. Vehicle detections from multiple perception sensors are integrated using track-to-track (T2T) fusion based on a cross-covariance method, and this T2T fusion is performed with curvilinear coordinates created using prebuilt roadway geometry on a digital map. The coordinate conversion to curvilinear space has many benefits for behavioral reasoning and tracking, such as constraining problem spaces and dimensions. A machine learning classifier based on a support vector machine is then applied to deduce the behavior of nearby vehicles. The algorithms presented here for track fusion and behavioral reasoning based on curvilinear coordinates have been verified through experiments in various real traffic scenarios.

Detection and tracking of moving objects is one of the most essential functions of autonomous cars. In order to estimate the dynamic information of a moving object accurately, laser scanners are widely used for their highly accurate distance data. However, these data only represent the surface of an object facing the sensor and changes the appearance of an object over time. This change produces unexpected tracking errors of estimated dynamic states. In this paper, in order to minimize the tracking error caused by appearance changes, a tracking algorithm based on L-shaped model switching is proposed. The suggested algorithm is validated in real traffic experiments where position, velocity, and heading angle error were measured by using precise GPS. The L-shape tracking algorithm successfully mitigated the effect of appearance changes and improved estimation performance.

This paper presents a hierarchical trajectory planning based on the integration of a sampling and an optimization method for urban autonomous driving. To manage a complex driving environment, the upper behavioral trajectory planner searches the macro-scale trajectory to determine the behavior of an autonomous car by using environment models, such as traffic control device and objects. This planner infers reasonable behavior and provides it to the motion trajectory planner. For planning the behavioral trajectory, the sampling-based approach is used due to its advantage of a free-form cost function for discrete models of the driving environments and simplification of the searching area. The lower motion trajectory planner determines the micro-scale trajectory based on the results of the upper trajectory planning with the environment model. The lower planner strongly considers vehicle dynamics within the planned behavior of the behavioral trajectory. Therefore, the planning space of the lower planner can be limited, allowing for improvement of the efficiency of the numerical optimization of the lower planner to find the best trajectory. For the motion trajectory planning, the numerical optimization is applied due to its advantages of a mathematical model for the continuous elements of the driving environments and low computation to converge minima in the convex function. The proposed algorithms of the sampling-based behavioral and optimization-based motion trajectory were evaluated through experiments in various scenarios of an urban area.

Tracking and behavior reasoning of surrounding vehicles on a roadway are keys for the development of automated vehicles and an advanced driver assistance system (ADAS). Based on dynamic information of the surrounding vehicles from the tracking algorithm and driver intentions from the behavior reasoning algorithm, the automated vehicles and ADAS can predict possible collisions and generate safe motion to avoid accidents. This paper presents a unified vehicle tracking and behavior reasoning algorithm that can simultaneously estimate the vehicle dynamic state and driver intentions. The multiple model filter based on various behavior models was used to classify the vehicle behavior and estimate the dynamic state of surrounding vehicles. In addition, roadway geometry constraints were applied to the unified vehicle tracking and behavior reasoning algorithm in order to improve the dynamic state estimation and the behavior classification performance. The curvilinear coordinate system was constructed based on the precise map information in order to apply the roadway geometry to the tracking and behavior reasoning algorithm. The proposed algorithm was verified and evaluated through experiments under various test scenarios. From the experimental results, we concluded that the presented tracking and behavior reasoning algorithm based on the roadway geometry constraints provides sufficient accuracy and reliability for automated vehicles and ADAS applications.

This paper proposes a road slope aided position estimation algorithm based on the fusion of GPS data with information from automotive onboard sensors. Many previous studies for position estimation did not consider the effect of road slope, although many sloped roads are existing. In order to analyze the influence of road slope on position estimation, theoretical proof and simulation are performed. Based on this analysis, a road slope aided position estimation algorithm is presented, which includes a vehicle motion model that can compensate for the effect of the road slope. This algorithm can estimate the position and road slope simultaneously. Furthermore, by compensating for the error due to the road slope, the algorithm can improve the position estimation accuracy and reliability. The estimation algorithm in this paper is implemented and evaluated using automotive onboard sensors and embedded system; therefore, additional motion sensors and high-performance computational units are not necessary. The experimental results show that the accuracy and reliability of the road slope aided position algorithm provide superior performance compared with a planar vehicle model-based position estimation algorithm in mountainous terrain.

This paper proposes a fast GPS and dead reckoning (DR) sensor fusion framework by applying a simplified coordinate conversion process to a Bayesian filter. To integrate the GPS and DR information, the two different coordinate systems of the GPS and DR must be unified. Many previous studies of coordinate conversion have been conducted to unify the coordinates of GPS and DR. However, computational limitations arise due to their complex conversion equations. Therefore, we present a simplified coordinate conversion process that approximately converts the DR coordinates into GPS coordinates based on linearization of the Earth geodetic model. By applying the simplified coordinate conversion to the GPS-DR fusion framework, we reduce the computation burden while remaining within the necessary tolerance for conversion error. The computational efficiency of the proposed framework is verified through experiments and comparisons with previous studies.

This paper presents a Monte Carlo localization algorithm for an autonomous car based on an integration of multiple sensors data. The sensor system is composed of onboard motion sensors, a low-cost GPS receiver, a precise digital map, and multiple cameras. Data from the onboard motion sensors, such as yaw rate and wheel speeds, are used to predict the vehicle motion, and the GPS receiver is applied to establish the validation boundary of the ego-vehicle position. The digital map contains location information at the centimeter level about road surface markers (RSMs), such as lane markers, stop lines, and traffic sign markers. The multiple images from the front and rear mono-cameras and the around-view monitoring system are used to detect the RSM features. The localization algorithm updates the measurements by matching the RSM features from the cameras to the digital map based on a particle filter. Because the particle filter updates the measurements based on a probabilistic sensor model, the exact probabilistic modeling of sensor noise is a key factor to enhance the localization performance. To design the probabilistic noise model of the RSM features more explicitly, we analyze the results of the RSM feature detection for various real driving conditions. The proposed localization algorithm is verified and evaluated through experiments under various test scenarios and configurations. From the experimental results, we conclude that the presented localization algorithm based on the probabilistic noise model of RSM features provides sufficient accuracy and reliability for autonomous driving system applications.

This paper presents a novel curvilinear-coordinate-based approach to improve object and situation assessment performance for highly automated vehicles under various curved road conditions. The approach integrates object information from radars and lane information from a camera with three steps: track-to-track fusion, curvilinear coordinate conversion, and lane assessment. The track-to-track fusion is achieved through a nearest neighbor filter that updates the target state estimation and covariance with the nearest neighbor measurement, and a cross-covariance method that merges the duplicate tracks using error covariance. In order to determine in which lane the fused tracks are located accurately and reliably, the curvilinear coordinate conversion process is performed. The curvilinear coordinates are generated in the form of a cubic Hermite spline lane model from the lane information of the camera. Based on the converted track information and the lane model in the curvilinear coordinates, the probability distribution of the threat levels in each lane is determined though a probabilistic lane association and threat assessment. The developed algorithm is verified and evaluated through experiments using a real-time embedded system. The results show that the proposed curvilinear-coordinate-based approach provides excellent performance of object and situation assessment, in respect of accuracy and computational efficiency, in real-time operation.

This paper proposes a map generation algorithm for a precise roadway map designed for autonomous cars. The roadway map generation algorithm is composed of three steps, namely, data acquisition, data processing, and road modeling. In the data acquisition step, raw trajectory and motion data for map generation are acquired through exploration using a probe vehicle equipped with GPS and on-board sensors. The data processing step then processes the acquired trajectory and motion data into roadway geometry data. GPS trajectory data are unsuitable for direct roadway map use by autonomous cars due to signal interruptions and multipath; therefore, motion information from the on-board sensors is applied to refine the GPS trajectory data. A fixed-interval optimal smoothing theory is used for a refinement algorithm that can improve the accuracy, continuity, and reliability of road geometry data. Refined road geometry data are represented into the B-spline road model. A gradual correction algorithm is proposed to accurately represent road geometry with a reduced amount of control parameters. The developed map generation algorithm is verified and evaluated through experimental studies under various road geometry conditions. The results show that the generated roadway map is sufficiently accurate and reliable to utilize for autonomous driving.

This paper presents a precise localization method for autonomous driving systems by correcting the GPS bias error. Since GPS errors have systematic noise properties that change slowly with time, a stand-alone GPS cannot be used for localization of an autonomous vehicle. To compensate for this systematic bias error, several types of additional sources of information, including on-board motion sensors, camera vision systems, and a road map database, are applied to the localization system. The localization algorithm is based on a particle filter, because the measurement model related to the representation of the road geometry is described by a highly nonlinear function. The proposed localization algorithm was tested and verified through an autonomous driving test.

This paper proposes a road-slope estimation algorithm to improve the performance and efficiency of intelligent vehicles. The algorithm integrates three types of road-slope measurements from a GPS receiver, automotive onboard sensors, and a longitudinal vehicle model. The measurement integration is achieved through a probabilistic data association filter (PDAF) that combines multiple measurements into a single measurement update by assigning statistical probability to each measurement and by removing faulty measurement via the false-alarm function of the PDAF. In addition to the PDAF, an interacting multiple-model filter (IMMF) approach is applied to the slope estimation algorithm to allow adaptation to various slope conditions. The model set of the IMMF is composed of a constant-slope road model (CSRM) and a constant-rate slope road model (CRSRM). The CSRM assumes that the slope of the road is always constant, and the CRSRM assumes that the slope of the road changes at a constant rate. The IMMF adapts the road-slope model to the driving conditions. The developed algorithm is verified and evaluated through experimental and case studies using a real-time embedded system. The results show that the performance and efficiency of the road-slope estimation algorithm is accurate and reliable enough for intelligent vehicle applications.

Vehicle position estimation for intelligent vehicles requires not only highly accurate position information but reliable and continuous information provision as well. A low-cost Global Positioning System (GPS) receiver has widely been used for conventional automotive applications, but it does not guarantee accuracy, reliability, or continuity of position data when GPS errors occur. To mitigate GPS errors, numerous Bayesian filters based on sensor fusion algorithms have been studied. The estimation performance of Bayesian filters primarily relies on the choice of process model. For this reason, the change in vehicle dynamics with driving conditions should be addressed in the process model of the Bayesian filters. This paper presents a positioning algorithm based on an interacting multiple model (IMM) filter that integrates low-cost GPS and in-vehicle sensors to adapt the vehicle model to various driving conditions. The model set of the IMM filter is composed of a kinematic vehicle model and a dynamic vehicle model. The algorithm developed in this paper is verified via intensive simulation and evaluated through experimentation with a real-time embedded system. Experimental results show that the performance of the positioning system is accurate and reliable under a wide range of driving conditions.

In the above titled paper (ibid., vol. 13, no. 1, pp. 329-343, March 2012), a production error occurred that resulted in the misspelling of the name of one of the authors of the paper. Keounyup Chu should be spelled Keonyup Chu. We regret the publication of this error.

A vehicle localization system can be extremely useful for intelligent transformation systems (ITS) such as advanced driver assistance systems (ADASs), emergency vehicle notification systems, and collision avoidance systems. To optimize the performance of vehicle localization systems, localization algorithms that analyze multi-sensor data processed using a Kalman filter have been developed. However, a Kalman filter with a single process model cannot guarantee the accuracy of localization under various driving conditions, because the single vehicle model does not cover all driving situations. Therefore, we present a position estimation algorithm based on an interacting multiple model (IMM) filter that uses two kinds of vehicle models: a kinematic vehicle model and a dynamic vehicle model. While the kinematic vehicle model is suitable for low-speed and low-slip driving conditions, the dynamic vehicle model is more appropriate for high-speed and high-slip situations. The IMM filter integrates the estimates from a kinematic vehicle model based on an extended Kalman filter (EKF) and estimates from a dynamic vehicle model based on EKF to improve localization accuracy. The developed estimation algorithm was verified by simulation using a commercial vehicle model. The simulation results show that the estimates of vehicle position by the algorithm presented in this study are accurate under a wide range of driving conditions.