Shankar C Subramanian

In this paper, a collision avoidance algorithm (CAA) has been proposed using variable time headway considering heterogeneous traffic. The time headway used in the proposed CAA was tuned based on the traffic scenarios, the host vehicle's load conditions and the type of the lead vehicle that the host vehicle encounters in the traffic. The proposed variable time headway would help to avoid the intervention of the collision avoidance system during normal driving and gain driver's acceptance. The CAA was evaluated using a hardware-in-the-loop (HiL) experimental set-up integrated with the vehicle dynamic simulation software IPG/TruckMaker® for different categories of lead vehicles such as 2/3 wheelers, passenger cars, light commercial road vehicles (LCVs) and heavy commercial road vehicles (HCVs). From the results, it was observed that while following a HCV, a smaller time headway was sufficient to prevent a collision compared to following a passenger car, LCV and 2/3 wheeler.

In this paper, a non-linear controller for collision avoidance in a heavy commercial road vehicle has been developed using Lyapunov theory. This paper considers the longitudinal dynamics of the vehicle, including the aerodynamic effect, the rolling resistance and the road grade. This paper also considers the maximum tire-road adhesion capacity and the braking capability of the vehicle. The developed controller has been tested using simulation for three realistic scenarios for different road and loading conditions and the results were compared with a controller developed using a linear full state feedback controller. It was observed that the non-linear controller has an advantage in terms of reduced time headway.

Advanced Driver Assistance Systems in heavy commercial road vehicles (HCRVs) have become necessary with the increasing contribution of HCRVs to road accident fatalities, where human factors form the major contributor. Collision avoidance systems can help in reducing fatalities through automated braking when required. This study focused on the development of a framework for the computation of the minimum time headway, for a Vehicle-to-Vehicle (V2V) communication-based collision avoidance system for HCRVs, that must be maintained through automated braking to avoid a collision. The framework, as well as the developed Collision Avoidance Algorithm (CAA), considers critical attributes of an HCRV such as significant load variations and brake actuator dynamics. Further, the time headway formulation varies with the initial longitudinal speed and considers communication latency. Experiments were conducted in a Hardware-in-Loop test setup to evaluate the efficacy of the proposed formulations. It was observed that the incorporation of communication latency and initial longitudinal speed explicitly in the time headway formulation maintained the desired final spacing between the vehicles, over a range of communication latencies and various host vehicle longitudinal speeds, which was not the case when the above factors were not considered.

An important aspect from the perspective of operational safety of heavy road vehicles is the detection and avoidance of collisions, particularly at high speeds. The development of a collision avoidance system is the overall focus of the research presented in this paper. The collision avoidance algorithm was developed using a sliding mode controller (SMC) and compared to one developed using linear full state feedback in terms of performance and controller effort. Important dynamic characteristics such as load transfer during braking, tyre-road interaction, dynamic brake force distribution and pneumatic brake system response were considered. The effect of aerodynamic drag on the controller performance was also studied. The developed control algorithms have been implemented on a Hardware-in-Loop experimental set-up equipped with the vehicle dynamic simulation software, IPG/TruckMaker®. The evaluation has been performed for realistic traffic scenarios with different loading and road conditions. The Hardware-in-Loop experimental results showed that the SMC and full state feedback controller were able to prevent the collision. However, when the discrepancies in the form of parametric variations were included, the SMC provided better results in terms of reduced stopping distance and lower controller effort compared to the full state feedback controller.