<p><i>In cooperative perception, reliable detection and localization of surrounding objects and communicating the information between vehicles is necessary for safety. &nbsp;Dynamic object detection and its trajectory tracking and transfer between v2v is a growing field of research. Especially, in a scenario where the object is not in the field of view of the ego vehicle, dynamically detecting the object and transferring its location in real-time to the ego vehicle will improve safety in autonomous conditions. To address this issue, in the current study, we have tested the reliability and ease of data tracking and transfer comparing range-based and vision-based sensors. &nbsp;Especially, transferring image data can be computationally expensive as compared to data point clouds while the reliability and localization of the dynamic object are of interest too in the current study. So, multiple scenarios have been created and tested where the dynamic object is not in the field of view of the ego vehicle.&nbsp; Imagine an intersection or vehicles on the same lane obstructing the view, in such a scenario, the lead vehicle needs to track the image or point clouds corresponding to the features of the object and send the data over v2v communication in real-time establishing cooperative perception. &nbsp;&nbsp;In connected vehicles, dynamic pose estimation and map merging between two Wifi-bots is necessary to ensure cooperative perception while the same can be simulated accurately on MATLAB or ROS environment. Traditionally, SIFT and SURF based algorithms have been used to match features between images from sensors looking at different perspectives to have a common field of view. The proposed study will incorporate novel pose estimation techniques and compare different feature extraction methods like MSER, ORB, FAST to test if better matches is available. &nbsp;&nbsp;Moreover, the dynamic object will have a linear trajectory in the current study which will be modified to non-linear and random trajectories in a future study. Overall, the study will address dynamic object tracking and trajectory transfer between two connected vehicles while comparing range and vision-based sensors and then test the accuracy, reliability, and ease of communication comparing both sensors. The research can be a precursor to future trajectory prediction of dynamic objects by the ego vehicle when there is a sudden loss of communication with the lead vehicle after the initial data transfer. Especially, future prediction of trajectory is necessary to incorporate path planning of ego vehicle to avoid collision with the surrounding object as well as the lead vehicle</i></p>

A Comparative Study on Feature Descriptors for Relative Pose Estimation in Connected Vehicles

Paper Proceedings: https://asmedigitalcollection.asme.org/IMECE/IMECE2021/volume/85543

ASME 2021 International Mechanical Engineering Congress and Exposition (IMECE2021)

Supporting the need among researchers worldwide for additional online and multimedia content, ASME now offers the ASME Conference Video Collection. This platform will collect and provide long-term access to conference presentation videos, encompassing all technology areas within ASME's technical conference and events program. About 30,000-40,000 engineers across all mechanical engineering and related disciplines participate in ASME conferences each year. This includes authors of approximately 6,000 to 8,000 technical papers published by ASME, the majority of whom also submit video presentations for our virtual conferences. The video collection will be available as an add-on to existing product packages or as a separate subscription.