Automotive Innovation’s Post

🔔🔔🔔 #MDPIfutureinternet [New Published Papers in 2024] Title: NeXtFusion: Attention-Based Camera-Radar Fusion Network for Improved Three-Dimensional Object Detection and Tracking Authors: Priyank Kalgaonkar and Mohamed El-Sharkawy Please read at: https://lnkd.in/gc2qsSq5 Keywords: #CondenseNeXt; sensor fusion; #objectdetection; #autonomousvehicle; PyTorch

By the way, what're the relationship between MAP & MAPE ? (2 mn) Few Week ago I shared an article written with some talented students (Anais Mongolo, Minnelli DE FONSECA PEIRIS) and a colleague (Alexis Capon). Within it, we proposed an original architecture for autonomous computer based on the MAPE architecture inspired from the autonomous vehicle area. Someone asked me if I should talk about MAP and explain why we choose MAPE in first place. The answer is quite simple MAP and MAPE are both architecture but not on the same subject. Fist of all, MAP stand for Modular Autonomous Platform although MAPE is for Monitor Analyse Plan and Execute. MAPE is a conceptual frame which allow to organize the system at a very high level with few principles just like the lateral symmetry of vertebrates in biology. MAP in its turn allow to organize the component of the system (molecularity, flexibility ... just as the endo skeleton vs exoskeleton or centralized neuronal system vs distributed neuronal system. So both architecture are compatible and complementary. We choose MAPE at first because it is one of the simplest conceptual frame for our project. 😁 What do you thing about these explanations ? Are you agree ?

🔔🔔🔔 #MDPIfutureinternet [New Published Papers in 2024] Title: NeXtFusion: Attention-Based Camera-Radar Fusion Network for Improved Three-Dimensional Object Detection and Tracking Authors: Priyank Kalgaonkar and Mohamed El-Sharkawy Please read at: https://lnkd.in/gCSiRqxB Keywords: #CondenseNeXt; sensor fusion; #objectdetection; #autonomousvehicle; PyTorch via Future Internet MDPI

Safe-Sim: Safety-Critical Closed-Loop Traffic Simulation with Diffusion-Controllable Adversaries Publication Date: 9/29/2024 Event: The 18th European Conference on Computer Vision #ECCV2024 Reference: pp. 1-17, 2024 Authors: Wei-Jer Chang, NEC Laboratories America, Inc., University of California, Berkeley; Francesco Pittaluga, NEC Laboratories America, Inc.; Masayoshi Tomizuka, University of California, Berkeley; Wei Zhan, University of California, Berkeley; Manmohan Chandraker, NEC Laboratories America, Inc. Abstract: Evaluating the performance of #autonomousvehicle planning algorithms necessitates simulating long-tail safety-critical traffic scenarios. However, traditional methods for generating such scenarios often fall short in terms of controllability and realism; they also neglect the dynamics of agent interactions. To address these limitations, we introduce Safe-Sim, a novel diffusion-based controllable closed-loop safety-critical simulation framework. Our approach yields two distinct advantages: 1) generating realistic long-tail safety-critical scenarios that closely reflect real-world conditions, and 2) providing controllable adversarial behavior for more comprehensive and interactive evaluations. We develop a novel approach to simulate safety-critical scenarios through an adversarial term in the denoising process of diffusion models, which allows an adversarial agent to challenge a planner with plausible maneuvers while all agents in the scene exhibit reactive and realistic behaviors. Furthermore, we propose novel guidance objectives and a partial diffusion process that enables users to control key aspects of the scenarios, such as the collision type and aggressiveness of the adversarial agent, while maintaining the realism of the behavior. We validate our framework empirically using the nuScenes and nuPlan datasets across multiple planners, demonstrating improvements in both realism and controllability. These findings affirm that diffusion models provide a robust and versatile foundation for safety-critical, interactive traffic simulation, extending their utility across the broader autonomous driving landscape. Learn more: https://lnkd.in/eGeRggcw

📢 Exciting News! 📢 I’m thrilled to share my latest research paper titled "An Enhanced Model for Detecting and Classifying Emergency Vehicles Using a Generative Adversarial Network (GAN)" has been published. In this work, we introduce a novel approach utilizing GANs to improve the accuracy and efficiency of detecting and classifying emergency vehicles, a crucial advancement for enhancing safety and response times. #Research #MachineLearning #GAN #EmergencyVehicles #Publication #ComputerVision

[white paper] [open access] Optimizing the image correction pipeline for pedestrian detection in the thermal-infrared domain Christophe Karam, Jessy Matias, Xavier Breniere, Jocelyn Chanussot https://lnkd.in/dmqAW6wB Infrared imagery can help in low-visibility situations such as fog and low-light scenarios, but it is prone to thermal noise and requires further processing and correction. This work studies the effect of different infrared processing pipelines on the performance of a pedestrian detection in an urban environment, similar to autonomous driving scenarios. Detection on infrared images is shown to outperform that on visible images, but the infrared correction pipeline is crucial since the models cannot extract information from raw infrared images. Two thermal correction pipelines are studied, the shutter and the shutterless pipes. Experiments show that some correction algorithms like spatial denoising are detrimental to performance even if they increase visual quality for a human observer. Other algorithms like destriping and, to a lesser extent, temporal denoising, increase computational time, but have some role to play in increasing detection accuracy. As it stands, the optimal trade-off for speed and accuracy is simply to use the shutterless pipe with a tonemapping algorithm only, for autonomous driving applications within varied environments. Fondation Grenoble INP LYNRED DATA SCIENCE EXPERTS #infrared #artificialintelligence

🎉 𝐄𝐱𝐜𝐢𝐭𝐢𝐧𝐠 𝐍𝐞𝐰𝐬! Our project on Advancing Robotic Terrain Classification Published by IEEE 📝 🔬Under Dr. Suddhasil De's guidance, Sudhanshu Tripathi, Sarvada Sakshi Jha, Shakti Deo Kumar and I developed a groundbreaking method for legged robots to navigate challenging terrains. 📈 🌟 Key Achievements: 💡 Overcame overfitting & low accuracy with innovative techniques. 🚀 Stacked LSTM architecture & unique loss regularization significantly boosted accuracy. 🌄 Achieved state-of-the-art results in real-time terrain classification. 🔍 Our robust performance was validated through rigorous comparative analysis. 🤖 Our robot can now figure out the ground in real-time better than ever before. This publication marks a significant milestone in our pursuit of advancing robotics technology! 🙌 #ieee #achievement #nitp #publication #robotic #terrain

Driving Futures: WcDT Autonomous Trajectory The paper introduces WcDT, a fusion of diffusion probabilistic models and transformers for autonomous driving trajectory generation. By leveraging historical data and HD map features, WcDT optimizes trajectory generation, resulting in realistic and diverse trajectories. Experimental results demonstrate its superior performance, indicating its potential for integration into automatic driving simulation systems, showcasing advancements in AI for autonomous vehicles. #WcDT #DiffusionProbabilisticModels #Transformers #AutonomousDriving #TrajectoryGeneration #HDMaps #AI #SimulationSystems #Advancements

💡 #Highcitedpaper #Dalian Maritime University 🌊 Title: Generalized Behavior Decision-Making Model for Ship Collision Avoidance via Reinforcement Learning Method 🔑 Keywords: #reinforcementlearning; multi-ship encounter situations; #collisionavoidance; #obstaclezone by target (#OZT); intelligent decision-making 🔗 paper link: https://lnkd.in/eX7U3Rsf 📜 Abstract：Due to the increasing number of transportation vessels, marine traffic has become more congested. According to the statistics, 89% to 95% of maritime accidents are related to human factors. In order to reduce marine incidents, ship automatic collision avoidance has become one of the most important research issues in the field of ocean engineering. A generalized behavior decision-making (GBDM) model, trained via a reinforcement learning (RL) algorithm, is proposed in this paper, and it can be used for ship autonomous driving in multi-ship encounter situations. Firstly, the obstacle zone by target (OZT) is used to calculate the area of future collisions based on the dynamic information of ships. Meanwhile, a virtual sensor called a grid sensor is taken as the input of the observation state. Then, International Regulations for Preventing Collision at Sea (COLREGs) is introduced into the reward function to make the decision-making fully comply with COLREGs. Different from the previous RL-based collision avoidance model, the interaction between the ship and the environment only works in the collision avoidance decision-making stage. Finally, 60 complex multi-ship encounter scenarios clustered by the COLREGs are taken as the ship’s GBDM model training environments. The simulation results show that the proposed GBDM model and training method has flexible scalability in solving the multi-ship collision avoidance problem complying with COLREGs in different scenarios.

How a human-inspired learning algorithm like Reinforcement Learning (RL) can be applied to minimize the fuel consumption of Hybrid Electric Vehicles (HEVs)? 🧠 🚗 If you are interested in understanding how to apply RL agents to the energy management of HEVs, let’s have a look at our latest article just published on Transportation Engineering by Elsevier, available at the following link: https://lnkd.in/dZGX3B-2 It is Open Access!! 🔓 The work shows a methodology to properly integrate a Deep Q-Learning agent into the Energy Management System (EMS) of a Plug-in Hybrid Electric Vehicle (PHEV). 🚗 🔋 We hope you'll find the article interesting, and we look forward to your comments! 🚀 #electrifiedpowertrain #reinforcementlearning #PHEV #RDE #EMS #MATLAB #Simulink #NumericalSimulation #transportation