The tradeoff between speed and accuracy is important in semantic segmentation problems, especially for the resource-constrained platforms, such as the intelligent vehicles. In this paper, we address this issue by proposing a well-deployed real-time semantic segmentation architecture named MLFNet. Specifically, we first build a lightweight backbone (SEFE) with a larger receptive field and multi-scale contextual representation performance to encode the pixel-level features. For better preserving target boundaries and contours, a spatial compensation branch (SPFE) is designed to gradually reduce the dimension of feature maps and refine the low-level specifics. In the decoding phase, we introduce a well-designed multi-branch fusion extractor (MBFD) for integrating the spatial details into high-level layers. Finally, the outputs from the semantic and spatial branches are fused to predict the final segmentation results. Extensive offline and online experiments have shown that our model has a superior speed and accuracy trade-off. On the Cityscapes test dataset, our model (MLFNet-Res18) achieves 71.0% mIoU with 95.1 FPS for 512 × 1024 inputs, and 72.1% mIoU with 72.2 FPS while inferring on MLFNet-Res34 model. Meanwhile, MLFNet-Res18 can reach 24.5 FPS when deployed to NVIDIA Jetson AGX Xavier and 64.0 FPS with an experimental vehicle.

This paper proposes a predictive cruise control based on eco-driving for a passage car that uses the information of upcoming traffic limits and the preceding vehicle to realize better fuel economy. To fully exploit the inherent potential of the powertrain system to reduce fuel consumption, the velocity is obtained by optimizing the engine torque, the brake force, and the gearshift while ensuring safe distance separation and traffic speed limits. The problem is described as a nonlinear mixed-integer problem and solved by the concept of combining Pontryagin's minimum principle and bisection method. The simulation results show a significant improvement in computational efficiency compared with traditional numerical methods, and the simulation results also show that the computational time increases linearly with prediction horizon. It is shown that an improvement of 8% in fuel is achieved in a realistic scenario compared with a basic vehicle using a standard adaptive cruise control.

The development of high-definition (HD) maps has enabled predictive cruise control (PCC) systems to access additional road and traffic information. This study provides a novel control scheme of PCC, which utilizes HD map information. To minimize fuel consumption, the problem of the PCC is formulated as a nonlinear model predictive control (MPC), and the derivation and implementation of the fast solver are discussed. Then, a novel shift-map is proposed to define the different working regions to allow the application of the proposed PCC system. The use of the real-time HD map is discussed, and the proposed control scheme is evaluated through simulation and experimental tests. The total fuel-savings rates obtained with the PCC system and factoryinstalled ACC system over a 370 km route were compared. An average fuel-savings rate of as high as 8.73% can be obtained by the proposed PCC system.

Guaranteed safety and performance under various circumstances remain technically critical and practically challenging for the wide deployment of autonomous vehicles. Safety-critical systems in general, require safe performance even during the reinforcement learning (RL) period. To address this issue, a Barrier Lyapunov Function-based safe RL (BLF-SRL) algorithm is proposed here for the formulated nonlinear system in strict-feedback form. This approach appropriately arranges and incorporates the BLF items into the optimized backstepping control method to constrain the state-variables in the designed safety region during learning. Wherein, thus, the optimal virtual/actual control in every backstepping subsystem is decomposed with BLF items and also with an adaptive uncertain item to be learned, which achieves safe exploration during the learning process. Then, the principle of Bellman optimality of continuous-time Hamilton–Jacobi–Bellman equation in every backstepping subsystem is satisfied with independently approximated actor and critic under the framework of actor-critic through the designed iterative updating. Eventually, the overall system control is optimized with the proposed BLF-SRL method. It is furthermore noteworthy that the variance of the attained control performance under uncertainty is also reduced with the proposed method. The effectiveness of the proposed method is verified with two motion control problems for autonomous vehicles through appropriate comparison simulations.