This paper presents a model for safe driving at blind intersections and its integration to a local planner based on a Frenet frame. The model predicts potential moving obstacles from blind intersections to proactively slow down to avoid potential collisions. The derivation of the model is described and its parameters are detailed. The local planner computes smooth trajectories with smooth velocity profiles so that the vehicle can follow the paths without jerk and sudden accelerations resulting in safe and comfortable navigation. Experimental results in simulation and in the real field with an autonomous car, show that the proposed predictive driving framework can reproduce human expert driver's trajectories and velocities when facing blind intersections.
This paper proposes gyro-offset and wheel radius estimation method using estimation results obtained by particle filter. Wheel radius and gyro-offset are important parameters of localization methods. One of solutions to estimate these parameters is to add these parameters as probabilistic variables to a probabilistic localization method. However, computation time of particle filter is depend on particle number and that number is incleased by degrees of freedom of parameters. This paper describes simple solutions of wheel radius and gyro-offset estimation based on particle filter and several experimental results.
This paper proposes robust localization method using free-space observation model based on particle filter. The proposed free-space observation model judges whether free-space of a laser beam of a sensor is proper. This paper describes the free-space observation model and several experimental results. In the experiments, the proposed localization algorithm achieves 1km localization in indoor and outdoor environments with many unkown objects.
A battery powered MAV is used to search inside damaged building and plumbing. However, its flight time is from 10 to 15 minutes because of the limitation of battery. We proposed a MAV has adhesion mechanism to improve above problem. It can hover longer time without propelling by using adhesion mechanism and winch mechanism. In this paper, we evaluate adhesion performance of the MAV in considerable environment.
In this work, we present a detailed comparison of ten different 3D LiDAR sensors, covering a range of manufacturers, models, and laser configurations, for the tasks of mapping and vehicle localization, using as common reference the Normal Distributions Transform (NDT) algorithm implemented in the self-driving open source platform Autoware. LiDAR data used in this study is a subset of our LiDAR Benchmarking and Reference (LIBRE) dataset, captured independently from each sensor, from a vehicle driven on public urban roads multiple times, at different times of the day. In this study, we analyze the performance and characteristics of each LiDAR for the tasks of (1) 3D mapping including an assessment map quality based on mean map entropy, and (2) 6-DOF localization using a ground truth reference map.
Due to the complexity of the environment and occlusions which often present in the scene, autonomous driving in an urban area is a challenging task. In some critical locations, e.g., intersections with low visibility, occlusions need to be taken into account as failing to do so might lead to a severe accident. In this paper, a method for crossing blind intersections with a mandatory stop using the estimated visibility of possible approaching vehicles is proposed. Speed profiles generated by the proposed method were compared with those of an expert driver. The results showed that the proposed method could produce a similar driving characteristic of an expert driver at low-visibility intersections with a mandatory stop.
