Cooper: Cooperative Perception for Connected Autonomous Vehicles Based on 3D Point Clouds

Abstract

Autonomous vehicles are heavily reliant upon their sensors to perfect the perception of surrounding environments, however, with the current state of technology, the data which a vehicle uses is confined to that from its own sensors. Data sharing between vehicles and/or edge servers is limited by the available network bandwidth and the stringent real-time constraints of autonomous driving applications. To address these issues, we propose a point cloud feature based cooperative perception framework (F-Cooper) for connected autonomous vehicles to achieve a better object detection precision. Not only will feature based data be sufficient for the training process, we also use the features' intrinsically small size to achieve real-time edge computing, without running the risk of congesting the network. Our experiment results show that by fusing features, we are able to achieve a better object detection result, around 10% improvement for detection within 20 meters and 30% for further distances, as well as achieve faster edge computing with a low communication delay, requiring 71 milliseconds in certain feature selections. To the best of our knowledge, we are the first to introduce feature-level data fusion to connected autonomous vehicles for the purpose of enhancing object detection and making realtime edge computing on inter-vehicle data feasible for autonomous vehicles.

F-cooper

“…Our F-Cooper framework supports two different fusion schemes: voxel feature fusion and spatial feature fusion. While the former achieves almost the same detection precision improvement when compared to the raw-data level fusion solution [3], the latter offers the ability to dynamically adjust the size of feature maps to be transmitted. A unique characteristic of F-Cooper is that it can be deployed and executed on in-vehicle and roadside edge systems.…”

“…That implies the simple fusion of object detection results from different cars would yield errors. Although fusing raw LiDAR data from two vehicles can improve the car detection precision [3], it is challenging to send the huge amount of LiDAR data generated by autonomous vehicles in real time. Solutions that increase vehicle's perception precision as well as maintaining or reducing computational complexity and turnaround time are rare in the literature.…”

Networked Roadside Perception Units for Autonomous Driving

“…Some researchers have also evaluated the realization of feature-level sensor data sharing in real vehicles [ 9 ]. In [ 10 ], the authors discussed the use of feature-level sensor data sharing to address the limited network bandwidth and stringent real-time constraints.…”

NeuralWalk: Trust Assessment in Online Social Networks with Neural Networks

“…Trust analysis is also introduced in cyber-physical systems (CPS), e.g., wireless sensor networks [88,89,90,91,92,93,94,95,96,97] and vehicular networks [98,99,100,101]. For example, a trust based framework is proposed to secure data aggregation in wireless sensor networks [102], which evaluates the trustworthiness of each sensor node by the Kullback-Leibler (KL) distance to identify the compromised nodes through an unsupervised learning technique.…”