With the increasing frequency of autonomous driving, more and more attention is paid to personalized path planning. However, the path selection preferences of users will change with internal or external factors. Therefore, this paper proposes a personalized path recommendation strategy that can track and study user’s path preference. First, we collect the data of the system, establish the relationship with the user preference factor, and get the user’s initial preference weight vector by dichotomizing the K‐means algorithm. The system then determines whether user preferences change based on a set threshold, and when the user’s preference changes, the current preference weight vector can be obtained by redefining the preference factor or calling difference perception. Finally, the road network is quantized separately according to the user preference weight vector, and the optimal path is obtained by using Tabu search algorithm. The simulation results of two scenarios show that the proposed strategy can meet the requirements of autopilot even when user preferences change.

The use of regenerative braking systems is an important approach for improving the travel mileage of electric vehicles, and the use of an auxiliary hydraulic braking energy recovery system can improve the efficiency of the braking energy recovery process. In this paper, we present an algorithm for optimizing the energy recovery efficiency of a hydraulic regenerative braking system (HRBS) based on fuzzy Q-Learning (FQL). First, we built a test bench, which was used to verify the accuracy of the hydraulic regenerative braking simulation model. Second, we combined the HRBS with the electric vehicle in ADVISOR. Third, we modified the regenerative braking control strategy by introducing the FQL algorithm and comparing it with a fuzzy-control-based energy recovery strategy. The simulation results showed that the power savings of the vehicle optimized by the FQL algorithm were improved by about 9.62% and 8.91% after 1015 cycles and under urban dynamometer driving schedule (UDDS) cycle conditions compared with a vehicle based on fuzzy control and the dynamic programming (DP) algorithm. The regenerative braking control strategy optimized by the fuzzy reinforcement learning method is more efficient in terms of energy recovery than the fuzzy control strategy.

In order to reduce the energy consumption caused by the frequent braking of vehicles at signalized intersections, an optimized speed trajectory control method is proposed, based on braking energy recovery efficiency (BERE) in connection with an automated system for vehicle real-time interaction with roadside facilities and regional central control. Our objectives were as follows; firstly, to establish the simulation model of the hybrid energy regenerative braking system (HERBS) and to verify it by bench test. Secondly, to build up the genetic algorithm (GA) optimization model for the deceleration stopping of the HERBS. Then, to obtain signal light status and timing information to be the constraints; the BERE is to be the optimized objective, resulting in optimization for the speed trajectory under the deceleration stopping condition of a single signalized intersection. Finally, vehicle simulations in ADVISOR software are utilized to validate the optimization results. The results show that the BERE during deceleration stopping at a single signalized intersection after the speed trajectory optimization is 36.21% higher than that of inexperienced drivers, and 7.82% higher than that of experienced drivers.

To address the shortcomings of previous autonomous decision models, which often overlook the personalized features of users, this paper proposes a personalized decision control algorithm for autonomous vehicles based on RLHF (reinforcement learning from human feedback). The algorithm combines two reinforcement learning approaches, DDPG (Deep Deterministic Policy Gradient) and PPO (proximal policy optimization), and divides the control scheme into three phases including pre-training, human evaluation, and parameter optimization. During the pre-training phase, an agent is trained using the DDPG algorithm. In the human evaluation phase, different trajectories generated by the DDPG-trained agent are scored by individuals with different styles, and the respective reward models are trained based on the trajectories. In the parameter optimization phase, the network parameters are updated using the PPO algorithm and the reward values given by the reward model to achieve personalized autonomous vehicle control. To validate the control algorithm designed in this paper, a simulation scenario was built using CARLA_0.9.13 software. The results demonstrate that the proposed algorithm can provide personalized decision control solutions for different styles of people, satisfying human needs while ensuring safety.

A six‐dimensional force sensor with a three‐beam structure was designed, and its strain and stress distribution under various load conditions were analyzed based on finite element simulation technology. According to the strain distribution, the Wheatstone bridge scheme of the sensor is designed, and the output voltage and sensitivity formulas in each direction are deduced. The optimal patch location was determined according to the strain distribution on the centerline of the strain beam. Finally, the size parameters of the elastic body were optimized based on the response surface method and multiobjective genetic algorithm, which greatly improved the sensitivity of the sensor in all directions. It can provide a useful guidance for the optimal design of the six‐dimensional force sensor with a three‐beam structure.