Study the Effect of Communication Delay for Perception and Collision Avoidance in Cooperative Autonomous Driving

The navigation function, collision mitigation, Adaptive Cruise Control (ACC), Cooperative Adaptive Cruise Control (CACC), and lane-keeping or changing; all these and more can be implemented from the earlier research as a resource for the development of technology related to the future autonomous driving. Modern radar, vision, and communication technology and their promising solutions have been analyzed from a prospect and dynamic barriers for large-scale as well as small-scale, which are dedicated to implementing the future fully autonomous on-road vehicles. The presence of GPS and wireless intercommunication networks inside the vehicle, one can inform autonomous vehicles, the surrounding vehicle situations, and the environmental conditions to compensate and also to anticipate all uncertain and critical obligations. Scientists and researchers are developing the enabling technologies to move the vehicles automatically for long-distance range with hazardous free roads. But, to move vehicles in all conditions on the road with uncertain disturbances, there must be some additional measurements and control capabilities called redundancy or stand-by in automation for all kinds of safety issues. It is a critical junction in the cooperative autonomous driving area where the goal is not only to improve the driving expertise in long-distance coverage. Also, to enhance driving safety and to prevent accidents caused by communication loss or delay due to environmental issues, and instantaneous uncertainties within short-distance coverage, especially within two to three times the braking distances.  Since a fully autonomous vehicle can only avoid hazards if the perception is cent percent accurate, and there is no communication delay or signal loss throughout the cooperative vehicle network. Hence the perception and communication are the two most essential tasks in autonomous driving, especially in a cooperative driving environment where the control of vehicle dynamics is not fully independent; rather, it is dependent on several instantaneous and other components of the network. Therefore, the proposed study focuses on a detailed analysis of communication signal loss or delay in a two-vehicle cooperative autonomous driving scenario on a simulation platform. The objective of which is a future stand-by solution to counter-act the delay and signal loss in autonomous driving.

The proposed study considers two on-road traffic scenarios in the MATLAB simulation platform to illustrate the real-time dynamic traffic environment and the fatality effect of communication loss for fully autonomous vehicles along with dynamic vulnerable road users such as a pedestrian. Few pedestrians follow the traffic rule and cross the road in an organized manner. Also, there are few pedestrians sometimes do not follow the traffic rule and cross the road randomly. Therefore, the autonomous vehicle should have the capability in their control algorithm to correct the proper distance and time gap and to take timely and appropriate control actions to avoid the collision or accident. Hence, the study considers an on-road traffic scenario, where two fully autonomous cooperative vehicles. One vehicle is known as ego vehicle, and the other is called lead vehicle. Both are capable enough to self-control their dynamics with an appropriate application of a longitudinal control algorithm to avoid an accident with a dynamic pedestrian. Though the effect of communication or transmission delay is considered in their control algorithms and if it is adequately estimated, then the correction is automatic, but it is almost impossible to predict the uncertain delay instantaneously. The proposed study will show that despite the cooperative perception and the casualty cannot be avoided or nullify if there is communication delay or signal loss between the autonomous vehicles. Therefore, the study will draw attention to take extra precaution in sensor redundancy as well as redundancy in the controllers unit to dynamically handle the specific or uncertain hazards which are otherwise unavoidable in a cooperative autonomous traffic environment.