arXiv subjects

Linbo Li

Publications and source records attributed to Linbo Li.

5 recordsLinked to original sources

arXiv2023

A reproducible approach to merging behavior analysis based on High Definition Map

Existing research on merging behavior generally prioritize the application of various algorithms, but often overlooks the fine-grained process and analysis of trajectories. This leads to the neglect of surrounding vehicle matching, the opaqueness of indicators definition, and reproducible crisis. To address these gaps, this paper presents a reproducible approach to merging behavior analysis. Specifically, we outline the causes of subjectivity and irreproducibility in existing studies. Thereafter, we employ lanelet2 High Definition (HD) map to construct a reproducible framework, that minimizes subjectivities, defines standardized indicators, identifies alongside vehicles, and divides scenarios. A comparative macroscopic and microscopic analysis is subsequently conducted. More importantly, this paper adheres to the Reproducible Research concept, providing all the source codes and reproduction instructions. Our results demonstrate that although scenarios with alongside vehicles occur in less than 6% of cases, their characteristics are significantly different from others, and these scenarios are often accompanied by high risk. This paper refines the understanding of merging behavior, raises awareness of reproducible studies, and serves as a watershed moment.

cs.RO↗

arXiv2021

Pareto-optimal lane-changing motion planning in mixed traffic

This paper applies the pareto-optimal concept to LC (lane-changing) motion planning in the presence of mixed traffic including manual and autonomous vehicles. Firstly, a multiobjective optimization problem is presented, in which the comfort, efficiency and safety of the LC vehicle and the surrounding vehicles are jointly modelled. Thereafter, the pareto-optimal solutions are obtained through employing the NSGA-II (Non-dominated Sorting Genetic -II) algorithm. Finally, the experiment section analyzes the (macroscopic and microscopic) lane-changing impact from a pareto-optimal perspective. Also, a comprehensive sensitivity analysis is conducted. Our results demonstrate that our algorithm could significantly reduce the lane-changing impact within its region, and the total costs are reduced in the range of 10.94% to 48.66%. This paper could be considered as a preliminary research framework for the application of the pareto-optimal concept. We hope this research will provide valuable insights into autonomous driving technology.

cs.RO↗

arXiv2021

Exploration of lane-changing duration for heavy vehicles and passenger cars: a survival analysis approach

Lane-changing (LC) behavior describes the lateral movement of the vehicle from the current-lane to the target-lane while proceeding forward. Among the many research directions, LC duration (LCD) measures the total time it takes for a vehicle to travel from the current lane to the target lane, which is an indispensable indicator to characterize the LC behavior. Although existing research has made some achievements, less attention has been paid to the research of heavy vehicles' LCD. Therefore, this paper aims to further explore the LCD between heavy vehicles and passenger cars. LC trajectories are extracted from the newly-released HighD dataset, which contains of 16.5 hours of measurement and over 11,000 vehicles. The survival function of LCD has been estimated, and the characteristic has been analyzed. Thereafter, the Accelerated Failure Time model is introduced to explore the influencing factors. Results demonstrate that the MST value of passenger cars and heavy vehicles is about 5.51s and 6.08s. The heavy vehicles would maintain a longer time-headway and distance-headway with preceding vehicle when performing LC. Nevertheless, these two factors do not significantly affect the LCD of heavy vehicles. Finally, the results and the modeling implications have been discussed. We hope this paper could contribute to our further understanding of the LC behaviors for heavy vehicles and passenger cars.

cs.RO↗

arXiv2021

Hierarchical automatic lane-changing motion planning: from self-optimum to local-optimum

In order to minimize the impact of lane change (LC) maneuver on surrounding traffic environment, a hierarchical automatic LC algorithm that could realize local optimum has been proposed. This algorithm consists of a tactical layer planner and an operational layer controller. The former generates a local-optimum trajectory. The comfort, efficiency, and safety of the LC vehicle and its surrounding vehicles are simultaneously satisfied in the optimization objective function. The later is designed based on vehicle kinematics model and the Model Predictive Control (MPC), which aims to minimize the tracking error and control increment. Combining macro-level and micro-level analysis, we verify the effectiveness of the proposed algorithm. Our results demonstrate that our proposed algorithm could greatly reduce the impact of LC maneuver on traffic flow. This is reflected in the decrease of total loss for nearby vehicles (such as discomfort and speed reduction), and the increase of traffic speed and throughput within the LC area. In addition, in order to guide the practical application of our algorithm, we employ the HighD dataset to validate the algorithm. This research could also be regarded as a preliminary foundational work to develop locally-optimal automatic LC algorithm. We anticipate that this research could provide valuable insights into autonomous driving technology.

cs.RO↗

arXiv2012

A Structured Construction of Optimal Measurement Matrix for Noiseless Compressed Sensing via Analog Polarization

In this paper, we propose a method of structured construction of the optimal measurement matrix for noiseless compressed sensing (CS), which achieves the minimum number of measurements which only needs to be as large as the sparsity of the signal itself to be recovered to guarantee almost error-free recovery, for sufficiently large dimension. To arrive at the results, we employ a duality between noiseless CS and analog coding across sparse additive noisy channel (SANC). Extending Renyi Information Dimension to Mutual Information Dimension (MID), we show the operational meaning of MID to be the fundamental limit of asymptotically error-free analog transmission across SANC under linear analog encoding constraint. We prove that MID polarizes after analog polar transformation and obeys the same recursive relationship as BEC. We further prove that analog polar encoding can achieve the fundamental limit of achievable dimension rate with vanishing Pe across SANC. From the duality, a structured construction scheme is proposed for the linear measurement matrix which achieves the minimum measurement requirement for noiseless CS.

cs.IT↗