Localization within high definition maps is a key problem for autonomous navigation as vehicles need to extract information from them. In addition, many navigation tasks are defined with respect to map features. For instance, estimating the cross-track and along-track gaps of a vehicle with respect to a given path is critical for lane keeping or intersection management. Map-based localization is also important for cooperative tasks like platooning in curved roads or lane changing. This work studies different methods to compute map-based coordinates defined as curvilinear abscissa, lateral distance and heading with respect to paths in high definition maps. Four approaches using polylines, lanelets and splines are compared. Thanks to real experiments, the discontinuity issues of polylines used in current high definition maps are evaluated and we discuss advantages and drawbacks of splines-based and lanelet methods. We also report experimental results corresponding to a platoon of two vehicles in curved roads and evaluate the effects of the use of low cost GNSS receivers.

This paper presents the complete system architecture of a connected driverless electric car designed to participate in the Grand Cooperative Driving Challenge 2016. One of the main goals of this challenge was to demonstrate the feasibility of multiple autonomous vehicles cooperating via wireless communications on public roads. Several complex cooperative scenarios were considered, including the merging of two lanes and cooperation at an intersection. We describe in some detail an implementation using the open-source PACPUS framework that successfully completed the different tasks in the challenge. Our description covers localization, mapping, perception, control, communication and the human-machine interface. Some experimental results recorded in real-time during the challenge are reported.

To navigate autonomously, a vehicle must be able to localize itself with respect to its driving environment and the vehicles with which it interacts. This work presents a decentralized cooperative localization method. It is based on the exchange of Local Dynamic Maps (LDM), which are cyberphysical representations of the physical driving environment containing poses and kinematic information about nearby vehicles. An LDM acts as an abstraction layer that makes the cooperation framework sensor-agnostic, and it can even improve the localization of a sensorless communicating vehicle. With this goal in mind, this work focuses on the property of consistency in LDM estimates. Uncertainty in the estimates needs to be properly modeled, so that the estimation error can be statistically bounded for a given confidence level. To obtain a consistent system, we first introduce a decentralized fusion framework that can cope with LDMs whose errors have an unknown degree of correlation. Second, we present a consistent method for estimating the relative pose between vehicles, using a 2D LiDAR with a point-to-line metric within an iterative-closest-point approach, combined with communicated polygonal shape models. Finally, we add a bias estimator in order to reduce position errors when non-differential GNSS receivers are used, based on visual observations of features geo-referenced in a High-Definition (HD) map. Real experiments were conducted, and the consistency of our approach was demonstrated on a platooning scenario using two experimental vehicles. The full experimental dataset used in this work is publicly available. A Poses transformations operatorsAs we use only direct orthonormal frames, the definition of one frame in its reference frame is equivalent to the frame transformation between these two frames and to the pose of this frame in its reference frame. The position corresponds to the origin and the orientation to the one of the x axes of the frame.

This work describes a cooperative pose estimation solution where several vehicles can perceive each other and share a geometrical model of their shape via wireless communication. We describe two formulations of the cooperation. In one case, a vehicle estimates its global pose from the one of a neighbor vehicle by localizing it in its body frame. In the other case, a vehicle uses its own pose and its perception to help localizing another one. An iterative minimization approach is used to compute the relative pose between the two vehicles by using a LiDAR-based perception method and a shared polygonal geometric model of the vehicles. This study shows how to obtain an observation of the pose of one vehicle given the perception and the pose communicated by another one without any filtering to properly characterize the cooperative problem independently of any other sensor. Accuracy and consistency of the proposed approaches are evaluated on real data from on-road experiments. It is shown that this kind of strategy for cooperative pose estimation can be accurate. We also analyze the advantages and drawbacks of the two approaches on a simple case study.