Object detection is a significant downstream task in computer vision. For the on-board edge computing platforms, a giant model is difficult to achieve the real-time detection requirement. And, a lightweight model built from a large number of the depth-wise separable convolution layers cannot achieve the sufficient accuracy. We introduce a new lightweight convolution technique, GSConv, to lighten the model but maintain the accuracy. The GSConv accomplishes an excellent trade-off between the model's accuracy and speed. And, we provide a design paradigm, slim-neck, to achieve a higher computational cost-effectiveness of the detectors. The effectiveness of our approach was robustly demonstrated in over twenty sets comparative experiments. In particular, the detectors of ameliorated by our approach obtains state-of-the-art results (e.g. 70.9% mAP0.5 for the SODA10M at a speed of ~ 100FPS on a Tesla T4 GPU) compared with the originals. Code is available at https://github.com/alanli1997/slim-neck-by-gsconv

In recent years, the increasing production and sales of automobiles have led to a notable rise in congestion on urban road traffic systems, particularly at ramps and intersections with traffic signals. Intelligent traffic signal control represents an effective means of addressing traffic congestion. Reinforcement learning methods have demonstrated considerable potential for addressing complex traffic signal control problems with multidimensional states and actions. In this research, the team propose Q-learning and Deep Q-Network (DQN) based signal control frameworks that use variable phase sequences and cycle times to adjust the order and the duration of signal phases to obtain a stable traffic signal control strategy. Experiments are simulated using the traffic simulator Simulation of Urban Mobility (SUMO) to test the average speed and the lane occupancy rate of vehicles entering the ramp to evaluate its safety performance and test the vehicle’s traveling time to assess its stability. The simulation results show that both reinforcement learning algorithms are able to control cars in dynamic traffic environments with higher average speed and lower lane occupancy rate than the no-control method and that the DQN control model improves the average speed by about 10% and reduces the lane occupancy rate by about 30% compared to the Q-learning control model, providing a higher safety performance.

Noise-vibration-harshness (NVH) design optimization problems have become major concerns in the vehicle product development process. The Body-in-White (BIW) plays an important role in determining the dynamic characteristics of vehicle system during the concept design phase. Finite Element (FE) models are commonly used for vehicle design. However, even though the speed of computers has been increased a lot, the simulation of FE models is still too time-consuming due to the increase in model complexity. For complex systems, like vehicle body structures, the numerous design variables and constraints make the FE simulations based optimization design inefficient. This calls for the development of a systematic and efficient approach that can effectively perform optimization to further improve the NVH performance, while satisfying the stringent design constraints.In the present work, an efficient method to optimize the structural dynamic response is proposed considering the low-frequency NVH performances. As a first step, to reduce computational burden, a response sensitivity analysis is performed to detect the most important variables prior to the design optimization. Then an analytical approximation model of vibration resonance peak is constructed and coupled with the adaptive simulated annealing (ASA) algorithm to replace the time-consuming finite element analysis. Subsequently, an optimization of NVH performance considering dynamic response is formulated and carried out. The methodology aims at improving the NVH behavior of body structure by simultaneously suppressing several resonance peaks. Finally, the proposed method and its process are successfully illustrated through a vehicle body example. The results demonstrate that the proposed method of incorporating response surface model with ASA algorithm is feasible and cost-efficient in solving the vibration optimization problem.