Aerodynamic integration produces a vehicle shape with a negative drag coefficient

Negative drag coefficients are normally associated with a vessel outfitted with a sail to extract energy from the wind and propel the vehicle forward. Therefore, the notion of a heavy vehicle, that is, a semi truck, that generates negative aerodynamic drag without a sail or any external appendages may seem implausible, especially given the fact that these vehicles have some of the largest drag coefficients on the road today. However, using both wind tunnel measurements and computational fluid dynamics simulations, we demonstrate aerodynamically integrated vehicle shapes that generate negative body-axis drag in a crosswind as a result of large negative frontal pressures that effectively “pull” the vehicle forward against the wind, much like a sailboat. While negative body-axis drag exists only for wind yaw angles above a certain analytical threshold, the negative frontal pressures exist at smaller yaw angles and subsequently produce body-axis drag coefficients that are significantly less than those of modern heavy vehicles. The application of this aerodynamic phenomenon to the heavy vehicle industry would produce sizable reductions in petroleum use throughout the United States.

Mono3D++: Monocular 3D Vehicle Detection with Two-Scale 3D Hypotheses and Task Priors

We present a method to infer 3D pose and shape of vehicles from a single image. To tackle this ill-posed problem, we optimize two-scale projection consistency between the generated 3D hypotheses and their 2D pseudo-measurements. Specifically, we use a morphable wireframe model to generate a fine-scaled representation of vehicle shape and pose. To reduce its sensitivity to 2D landmarks, we jointly model the 3D bounding box as a coarse representation which improves robustness. We also integrate three task priors, including unsupervised monocular depth, a ground plane constraint as well as vehicle shape priors, with forward projection errors into an overall energy function.

Driving Together Across Vehicle

Most research on vehicle-to-vehicle (V2V) communication is technology-driven, or focused on driver-to-driver interaction. Social communication between drivers and passengers across vehicles, with the same destination, is often neglected. Communication is influenced by context and occupant behavior, and has a significant effect on the collaborative driving scenario. An exploratory in-situ study with seven groups of two driver/co-driver pairs each, located in two separate vehicles, was conducted. On a predefined route, different subtasks had to be solved in a collaborative way. The study revealed a significant influence of different social factors, such as driving behavior, and contextual factors such as weather conditions, or vehicle shape and size. Findings delivered important insights and a deeper understanding on collaborative driving that may influence future V2V communication technologies. Additionally, the collaborative driving behavior of the driver/co-driver pairs could be transferred to a multi-agent framework.