--In this paper, we propose a search-based interactive motion planning scheme for autonomous vehicles (A Vs), using a game-theoretic approach. In contrast to traditional search-based approaches, the newly developed approach considers other road users (e.g. This leads to the generation of a more realistic path for the A V . Due to the low computational time, the proposed motion planning scheme is implementable in real-time applications. The performance of the developed motion planning scheme is compared with existing motion planning techniques and validated through experiments using W A T onoBus, an electrical all-weather autonomous shuttle bus. NTELLIGENT vehicles have increased their capabilities for highly automated driving under controlled environments i.e., driving scenarios that are designed to be predictable, stable, and safe for autonomous vehicles (A Vs) to operate in [1], [2]. Scene information is received using onboard sensors and communication network systems, i.e., infrastructure and other vehicles. Considering the available information, different motion planning and control techniques have been developed for autonomously driving in complex environments. The main goal is focused on executing strategies to improve safety, comfort, and energy optimization. One of the essential conditions for A V safety is ensuring safe interactions with other road users, including human-driven vehicles as well as pedestrians.

The integration of autonomous vehicles (AVs) into transportation systems presents an unprecedented opportunity to enhance road safety and efficiency. However, understanding the interactions between AVs and human-driven vehicles (HVs) at intersections remains an open research question. This study aims to bridge this gap by examining behavioral differences and adaptations of AVs and HVs at unsignalized intersections by utilizing two comprehensive AV datasets from Waymo and Lyft. Using a systematic methodology, the research identifies and analyzes merging and crossing conflicts by calculating key safety and efficiency metrics, including time to collision (TTC), post-encroachment time (PET), maximum required deceleration (MRD), time advantage (TA), and speed and acceleration profiles. The findings reveal a paradox in mixed traffic flow: while AVs maintain larger safety margins, their conservative behavior can lead to unexpected situations for human drivers, potentially causing unsafe conditions. From a performance point of view, human drivers exhibit more consistent behavior when interacting with AVs versus other HVs, suggesting AVs may contribute to harmonizing traffic flow patterns. Moreover, notable differences were observed between Waymo and Lyft vehicles, which highlights the importance of considering manufacturer-specific AV behaviors in traffic modeling and management strategies for the safe integration of AVs. The processed dataset utilized in this study is openly published to foster the research on AV-HV interactions.

The year had started so well for robotaxis. Cruise and Waymo came into 2023 riding high on fresh investments from General Motors and Google, respectively, as well as rapidly growing interest from the general public and a downright rabid rate of adoption by city governments. Things were looking up, very up, for the burgeoning self-driving vehicle industry! Then a driverless Crusie taxi accidentally dragged a hit-and-run victim down a San Francisco street for a few dozen feet and everything just sort of went to shit from there. Let's take a look back through the year that was to see how autonomous taxi tech might recover from this catastrophe.

Car-Following (CF), as a fundamental driving behaviour, has significant influences on the safety and efficiency of traffic flow. Investigating how human drivers react differently when following autonomous vs. human-driven vehicles (HV) is thus critical for mixed traffic flow. Research in this field can be expedited with trajectory datasets collected by Autonomous Vehicles (AVs). However, trajectories collected by AVs are noisy and not readily applicable for studying CF behaviour. This paper extracts and enhances two categories of CF data, HV-following-AV (H-A) and HV-following-HV (H-H), from the open Lyft level-5 dataset. First, CF pairs are selected based on specific rules. Next, the quality of raw data is assessed by anomaly analysis. Then, the raw CF data is corrected and enhanced via motion planning, Kalman filtering, and wavelet denoising. As a result, 29k+ H-A and 42k+ H-H car-following segments are obtained, with a total driving distance of 150k+ km. A diversity assessment shows that the processed data cover complete CF regimes for calibrating CF models. This open and ready-to-use dataset provides the opportunity to investigate the CF behaviours of following AVs vs. HVs from real-world data. It can further facilitate studies on exploring the impact of AVs on mixed urban traffic.

Deep Reinforcement Learning (DRL) enhances the efficiency of Autonomous Vehicles (AV), but also makes them susceptible to backdoor attacks that can result in traffic congestion or collisions. Backdoor functionality is typically incorporated by contaminating training datasets with covert malicious data to maintain high precision on genuine inputs while inducing the desired (malicious) outputs for specific inputs chosen by adversaries. Current defenses against backdoors mainly focus on image classification using image-based features, which cannot be readily transferred to the regression task of DRL-based AV controllers since the inputs are continuous sensor data, i.e., the combinations of velocity and distance of AV and its surrounding vehicles. Our proposed method adds well-designed noise to the input to neutralize backdoors. The approach involves learning an optimal smoothing (noise) distribution to preserve the normal functionality of genuine inputs while neutralizing backdoors. By doing so, the resulting model is expected to be more resilient against backdoor attacks while maintaining high accuracy on genuine inputs. The effectiveness of the proposed method is verified on a simulated traffic system based on a microscopic traffic simulator, where experimental results showcase that the smoothed traffic controller can neutralize all trigger samples and maintain the performance of relieving traffic congestion

We propose a model predictive control approach for autonomous vehicles that exploits learned Gaussian processes for predicting human driving behavior. The proposed approach employs the uncertainty about the GP's prediction to achieve safety. A multi-mode predictive control approach considers the possible intentions of the human drivers. While the intentions are represented by different Gaussian processes, their probabilities foreseen in the observed behaviors are determined by a suitable online classification. Intentions below a certain probability threshold are neglected to improve performance. The proposed multi-mode model predictive control approach with Gaussian process regression support enables repeated feasibility and probabilistic constraint satisfaction with high probability. The approach is underlined in simulation, considering real-world measurements for training the Gaussian processes.

As autonomous vehicles (AVs) continue to be integrated into public roads, it is inevitable that they will interact with human-driven vehicles (HVs) in a mixed traffic environment. In such traffic scenarios, it is crucial to consider the reactive and uncertain behavior of HVs when developing control strategies for AVs. This paper investigates the safe control of a platoon of AVs interacting with HVs in longitudinal car-following scenarios. To better predict the behavior of HVs, we propose a model that combines a first-principles nominal model with a Gaussian process (GP) learning-based component. Our results show that this model reduces the root mean square error in predicting HV velocity by 35.64\% compared to the nominal model. Utilizing this model, a model predictive control (MPC) strategy, referred to as GP-MPC, is designed to ensure a safe distance between each vehicle in the mixed vehicle platoon. The GP-MPC integrates the uncertainty assessment of the human-driven vehicle model by the GP models into the distance constraint, which enhances safety guarantees in challenging traffic scenarios such as emergency braking. Simulation case studies comparing the proposed GP-MPC against a baseline MPC demonstrate that the GP-MPC achieves superior safety guarantees while enabling more efficient motion behaviors for all vehicles in the mixed vehicle platoon.

There's an old proverb that dates back to at least the year 1670 and declares that sauce for the goose is also sauce for the gander. A more modern and altogether familiar version is the assertion that what is good for the goose is good for the gander. That's a saying that we all know well. In today's world, this ostensibly suggests that something applicable in one instance is likely applicable in another (consult your favored online dictionary for further elaboration). I often highlight cutting-edge technology bringing about AI-based true self-driving cars. I like to highlight foundational R&D work taking place in research labs that are focused on creating autonomous vehicles. The thing is, a lot of the autonomous tech will also end up in human-driven cars too. Many assume that the tech devised to aid AI-based autonomous driving would solely be used by autonomous driving vehicles.

Bestmile has raised $16.5 million in venture capital as it continues to expand its platform for managing autonomous and human-driven vehicles. The Switzerland-based company's software was originally designed to help transportation companies optimize fleets of self-driving vehicles, including cars, trucks, and shuttles. But Bestmile has evolved to include traditional vehicles as the shift to an autonomous world is now stretching out over many years. "We are creating this kind of flexibility of the platform because the transition to fully autonomous will be maybe 15 years," said Bestmile cofounder and CEO Raphael Gindrat, "so our customers need one platform to manage both in parallel." Founded in 2014, the company now has more than 55 employees, most of whom are located in its Lausanne headquarters, with a handful in North America and in Asia. Bestmile currently has 15 customers and is particularly embraced by companies that have deployed autonomous shuttles in strictly defined zones.

Abstract-- In this paper, we describe a framework for autonomous decision making in a dynamic and interactive environment based on cognitive hierarchy theory. We model the in - teractions between the ego agent and its operating environm ent as a two-player dynamic game, and integrate cognitive behav - ioral models, Bayesian inference, and receding-horizon op timal control to define a dynamically-evolving decision strategy for the ego agent. Simulation examples representing autonomou s vehicle control in three traffic scenarios where the autonom ous ego vehicle interacts with a human-driven vehicle are repor ted. Autonomous systems are becoming more capable, better accepted, and more commonplace. Many autonomous systems, including collaborative robots [1] and self-driv ing cars [2], operate in dynamic and interactive environments.