Search

Due to the significant increase in the use of autonomous car technology, it is essential to integrate this technology with high-precision digital map data containing more precise and accurate roadway information, as compared to existing conventional map resources, to ensure the safety of self-driving operations. While existing map technologies may assist vehicles in identifying their locations via Global Positioning System, it is however difficult to update the environmental changes of roadways in these maps. Roadway vision algorithms can be useful for building autonomous vehicles that can avoid accidents and detect real-time location changes. We incorporate a hybrid architectural design that combines unsupervised classification of vision data with supervised joint fusion classification to achieve a better noise-resistant algorithm. We identify, via a deep learning approach, an intelligent hybrid fusion algorithm for fusing multimodal vision feature data for roadway classifications and characterize its improvement in accuracy over unsupervised identifications using image processing and supervised vision classifiers. We analyzed over 93,000 vision frame data collected from a test vehicle in real roadways. The performance indicators of the proposedhybrid fusion algorithm are successfully evaluated for the generation of roadway digital maps for autonomous vehicles, with a recall of 0.94, precision of 0.96, and accuracy of 0.92. [ABSTRACT FROM AUTHOR]

This paper describes the design and verification of a control algorithm for advanced lane-change assist systems with integrated control of the differential braking and the electric power steering system. The objective of the proposed control algorithm of the advanced lane-change assist system is to minimize the unexpected control input of electric power steering and to make maximum use of the differential brake effort. The advanced lane-change assist system can warn of potential rear and side crashes and prevent collisions in lane-change manoeuvres with active intervention when there is an upcoming vehicle with a potential risk to the rear in the next lane. An enhanced lane-change assist warning algorithm is developed with additional information on the estimated past trajectory of the subject vehicle. The upper controller of the advanced lane-change assist system determines the control on–off decision, the desired yaw rate for collision avoidance and the strategic control input distribution. The key strategy of this integrated control algorithm is to use the maximum tyre–road friction of the differential brake force and to operate with a smaller control effort of the electric power steering only when lacking in the level of yaw rate moment obtained by the electronic stability control, which is intended to minimize annoyance to a driver and control intervention. The lower controller decides the control input of the advanced electronic stability control and the electric power steering system. Finally the advanced lane-change assist system is implemented in a real vehicle and tested in both the steering-control-only case and the integrated-control case. It is shown that the proposed strategy can intervene appropriately and verifies the effectiveness of collision avoidance in a dangerous lane-change situation; a decrease in the electric power steering control input of about 20% can be achieved by the proposed algorithm. [ABSTRACT FROM PUBLISHER]