This paper presents a scientometric and bibliometric review of the research on autonomous vehicles (AVs) to identify its main characteristics, evolution, and potential trends for future studies. Relevant articles were searched on WoS, yielding a research corpus of 10,580 papers, and the software CiteSpace was subsequently used for analysis. The results showed that AV research is heterogeneous and registered a growing demand over time. Multidisciplinarity is present, with 96 science fields being identified. As in any other sector, it is necessary to understand broader aspects of this industry such as the market factors surrounding it, as well as other economic and managerial issues. In this sense, we observed a migration of the research field from multidisciplinarity to pluridisciplinarity with a greater number of studies focusing on the latter. We understand that terminology standardization contributes to achieving pluridisciplinarity. As such, it is important to highlight that sustainability, public policies, liability, and safety, as well as business issues such as performance and business models are some of the tendencies in the field of AVs. For future studies, we suggest a more in-depth analysis of publications in terms of individual search terms, as well as the sub-areas identified as trends in this paper.
This paper presents an approach for road detection based on image segmentation. This segmentation is resulted from merging 2D and 3D image processing data from a stereo vision system. The 2D layer returns a matrix containing pixel's clusters based on the Watershed transform. Whereas the 3D layer return labels, that are classified by the V-Disparity technique, to free spaces, obstacles and non-classified area. Thus, a feature's descriptor for each cluster is composed with features from both layers. The road pattern recognition was performed by an artificial neural network, trained to obtain a final result from this feature's descriptor. The proposed work reports real experiments carried out in a challenging urban environment to illustrate the validity and application of this approach.
This work presents a safe navigation approach for a carlike robot. The approach relies on a global motion planning based on Velocity Vector Fields along with a Dynamic Window Approach for avoiding unmodeled obstacles. Basically, the vector field is associated with a kinematic, feedback-linearization controller whose outputs are validated, and eventually modified, by the Dynamic Window Approach. Experiments with a full-size autonomous car equipped with a stereo camera show that the vehicle was able to track the vector field and avoid obstacles in its way.
This paper proposes a global navigation strategy for autonomous vehicle combining sensor based control and digital maps information. In order to avoid localization problems in urban environments, this approach intends to solve the global navigation focusing on two local navigation tasks: road lane following and road intersection maneuvers. For that, it is important to use digital maps, once they provide a rich database containing information about the environment structure, like speed limit, number of lanes, traffic directions, etc. Associating the data extracted from the digital map with local navigation approaches, the vehicle is able to perform its global navigation task. The experiments take into account real data and a simulated scenario, which show the viability of this approach.
International audienceThis paper presents a new hybrid control approach for vision-based navigation applied to autonomous robotic automobiles in urban environments. It is composed by a Visual Servoing (VS) for road lane following (as deliberative control) and a Dynamic Window Approach (DWA) for obstacle avoidance (as reactive control). Typically, VS applications do not change the velocities to stop the robot in dangerous situations or avoid obstacles while performing the navigation task. However, in several urban conditions, these are elements that must be dealt with to guarantee the safe movement of the car. As a solution for this problem, in this study a line following VS controller will be used to perform road lane following tasks with obstacle avoidance, validating its control outputs in a new Image-Based Dynamic Window Approach (IDWA). The final solution combines the benefits of both controllers (VS+IDWA) for optimal lane following and fast obstacle avoidance, taking into account the car kinematics and some dynamics constraints. Experiments in a challenging scenario with both simulated and real experimental car show the viability of the proposed methodology
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.