Recently, vehicle platooning has attracted a lot of attention as one of the promising solutions to the explosion of vehicle numbers. By exchanging information among vehicles through communication networks, vehicle platooning can significantly improve traffic safety and efficiency. In this paper, we develop a joint systematic design of platoon communication and control to reduce position errors of consecutive vehicles and to improve platoon safety. Through separate information dissemination of the platoon leader and followers, we improve the success probability of the platoon leader's information dissemination. To extend the communication range of the platoon leader and also the platoon scale, we propose to use relays to forward the platoon leader's messages. Further, an adaptive distributed model predictive control (DMPC) based controller is presented, which can adjust its control parameters according to platoon state and dynamic information sources. Simulation results not only verify the proposed systematic design in terms of position errors, but also show that our scheme performs well in vehicle failure cases where collisions can be avoided and platoon safety improved.

Depth Completion can produce a dense depth map from a sparse input and provide a more complete 3D description of the environment. Despite great progress made in depth completion, the sparsity of the input and low density of the ground truth still make this problem challenging. In this work, we propose DenseLiDAR, a novel real-time pseudo-depth guided depth completion neural network. We exploit dense pseudo-depth map obtained from simple morphological operations to guide the network in three aspects: (1) Constructing a residual structure for the output; (2) Rectifying the sparse input data; (3) Providing dense structural loss for training the network. Thanks to these novel designs, higher performance of the output could be achieved. In addition, two new metrics for better evaluating the quality of the predicted depth map are also presented. Extensive experiments on KITTI depth completion benchmark suggest that our model is able to achieve the state-of-the-art performance at the highest frame rate of 50 Hz. The predicted dense depth is further evaluated by several downstream robotic perception or positioning tasks. For the task of 3D object detection, 3~5 percent performance gains on small objects categories are achieved on KITTI 3D object detection dataset. For RGB-D SLAM, higher accuracy on vehicle's trajectory is also obtained in KITTI Odometry dataset. These promising results not only verify the high quality of our depth prediction, but also demonstrate the potential of improving the related downstream tasks by using depth completion results.

In this paper, we investigate the joint sub-channel and power allocation problem for cellular vehicle-to-everything (V2X) communications, where multiple vehicle-to-infrastructure (V2I) users share the spectrum resources with vehicle-to-vehicle (V2V) users. In particular, a novel channel state information (CSI)-independent decentralized algorithm based on multi-agent reinforcement learning (MARL) is proposed to maximize the sum throughput of V2I links while meeting the latency and reliability requirements of V2V links. Specifically, we implement the individual double dueling deep recurrent Q-networks (D3RQN) and the carefully designed common reward to train the implicitly collaborative agents, through which, each agent optimizes the policy individually based solely on local CSI-independent observations. To handle the non-stationarity induced by multi-agent concurrent learning, we incorporate hysteretic Q-learning and concurrent experience replay trajectory (CERT) to stabilize the training process. Besides, we incorporate the approximate regretted reward (ARR) to alleviate the unstable reward estimation problem caused by shifting environment dynamics. Simulation results demonstrate that the proposed algorithm outperforms the baselines and can achieve close performance compared with the centralized Brute-force method. Furthermore, the proposed CSI-independent design exhibits comparable performance as the CSI-involved version, which sheds some light on the potential to further reduce the signalling overhead of machine learning-based vehicular communication systems.

As a low-level task of 3D perception, scene flow is a fundamental representation of dynamic scenes and provides non-rigid motion descriptions for the objects in the 3D environment, which can strongly support many upper-level applications. Inspired by the revolutionary success of deep learning, many attention-based neural networks have recently been proposed to estimate scene flow from consecutive point clouds. However, extracting effective features and estimating accurate point motions for irregular and occluded point clouds remains a challenging task. In this letter, we propose PT-FlowNet, the first end-to-end scene flow estimation network embedding the point transformer (PT) into all functional stages of the task. In particular, we design novel PT-based modules for point feature extraction, iterative flow update, and flow refinement stage to encourage effective point-level feature aggregation. Experimental results on FlyingThings3D and KITTI datasets show that our PT-FlowNet achieves state-of-the-art performance. Trained on synthetic data only, our PT-FlowNet can generalize to real-world scans and outperforms the existing methods by at least 36.2% for the EPE3D metric on the KITTI dataset.

A polarization-based method for water hazards detection is presented. The concept of polarization is introduced to computer vision to detect water hazards for autonomous off-road navigation. This method is based on the physical principle that the light reflected from water surface is partial linearly polarized and the polarization phases of them are more similar than those from the scenes around. Water hazards can be detected by comparison of polarization degree and similarity of the polarization phases. Experiments show that the method has good performance in water detection in complex natural backgrounds, especially when there is vegetation reflected in the water. This method has significant complementary advantages with respect to existing techniques, is computationally efficient, and can be easily implemented with existing imaging technology.

As the core task of scene understanding, semantic segmentation and depth completion play a vital role in lots of applications such as robot navigation, AR/VR and autonomous driving. They are responsible for parsing scenes from the angle of semantics and geometry, respectively. While great progress has been made in both tasks through deep learning technologies, few works have been done on building a joint model by deeply exploring the inner relationship of the above tasks. In this paper, semantic segmentation and depth completion are jointly considered under a multi-task learning framework. By sharing a common encoder part and introducing boundary features as inner constraints in the decoder part, the two tasks can properly share the required information from each other. An extra boundary detection sub-task is responsible for providing the boundary features and constructing cross-task joint loss functions for network training. The entire network is implemented end-to-end and evaluated with both RGB and sparse depth input. Experiments conducted on synthesized and real scene datasets show that our proposed multi-task CNN model can effectively improve the performance of every single task.