V2P (vehicle-to-pedestrian) communication can improve road traffic efficiency, solve traffic congestion, and improve traffic safety. It is an important direction for the development of smart transportation in the future. Existing V2P communication systems are limited to the early warning of vehicles and pedestrians, and do not plan the trajectory of vehicles to achieve active collision avoidance. In order to reduce the adverse effects on vehicle comfort and economy caused by switching the “stop–go” state, this paper uses a PF (particle filter) to preprocess GPS (Global Positioning System) data to solve the problem of poor positioning accuracy. An obstacle avoidance trajectory-planning algorithm that meets the needs of vehicle path planning is proposed, which considers the constraints of the road environment and pedestrian travel. The algorithm improves the obstacle repulsion model of the artificial potential field method, and combines it with the A* algorithm and model predictive control. At the same time, it controls the input and output based on the artificial potential field method and vehicle motion constraints, so as to obtain the planned trajectory of the vehicle’s active obstacle avoidance. The test results show that the vehicle trajectory planned by the algorithm is relatively smooth, and the acceleration and steering angle change ranges are small. Based on ensuring safety, stability, and comfort in vehicle driving, this trajectory can effectively prevent collisions between vehicles and pedestrians and improve traffic efficiency.

In this paper, a lightweight, high-definition mapping method is proposed for autonomous driving to address the drawbacks of traditional mapping methods, such as high cost, low efficiency, and slow update frequency. The proposed method is based on multi-source data fusion perception and involves generating local semantic maps (LSMs) using multi-sensor fusion on a vehicle and uploading multiple LSMs of the same road section, obtained through crowdsourcing, to a cloud server. An improved, two-stage semantic alignment algorithm, based on the semantic generalized iterative closest point (GICP), was then used to optimize the multi-trajectories pose on the cloud. Finally, an improved density clustering algorithm was proposed to instantiate the aligned semantic elements and generate vector semantic maps to improve mapping efficiency. Experimental results demonstrated the accuracy of the proposed method, with a horizontal error within 20 cm, a vertical error within 50 cm, and an average map size of 40 Kb/Km. The proposed method meets the requirements of being high definition, low cost, lightweight, robust, and up-to-date for autonomous driving.

LoRa (long range) is a communication technology that employs chirp spread spectrum modulation. Among various low-power wide area network (LPWAN) technologies, LoRa offers unique advantages, including low power consumption, long transmission distance, strong anti-interference capability, and high network capacity. Addressing the issue of power-constrained end devices in IoT application scenarios, this paper proposes an adaptive data rate (ADR) algorithm for LoRa networks designed for power-constrained end devices (EDs). The algorithm evaluates the uplink communication link state between the EDs and the gateway (GW) by using a combined weighting method to comprehensively assess the signal-to-noise ratio (SNR), received signal strength indication (RSSI), and packet reception rate (PRR), and calculates a list of transmission power and data rates that ensure stable and reliable communication between the EDs and the GW. By using ED power consumption models, network throughput models, and ED latency models to evaluate network performance, the Zebra optimization algorithm is employed to find the optimal data rate for each ED under power-constrained conditions while maximizing network performance. Test results show that, in a single ED scenario, the average PRR achieved by the proposed ADR algorithm for power-constrained EDs in LoRa networks is 14% higher than that of the standard LoRaWAN ADR algorithm. In a multi-ED link scenario (50 end devices), the proposed method reduces the average power consumption of EDs by 10% compared to LoRaWAN ADR, achieves a network throughput of 6683 bps, and an average latency of 2.10 s, demonstrating superior performance overall. The proposed method shows unique advantages in LoRa networks with power-constrained EDs and a large number of EDs, as it not only reduces the average power consumption of the EDs but also optimizes network throughput and average latency.