To plan a safe and efficient route, an autonomous vehicle should anticipate future motions of other agents around it. Motion prediction is an extremely challenging task that recently gained significant attention within the research community. In this work, we present a simple and yet very strong baseline for multimodal motion prediction based purely on Convolutional Neural Networks. While being easy-to-implement, the proposed approach achieves competitive performance compared to the state-of-the-art methods and ranks 3rd on the 2021 Waymo Open Dataset Motion Prediction Challenge. Our source code is publicly available at GitHub
相關內容
This paper presents a multi-layer motion planning and control architecture for autonomous racing, capable of avoiding static obstacles, performing active overtakes, and reaching velocities above 75 $m/s$. The used offline global trajectory generation and the online model predictive controller are highly based on optimization and dynamic models of the vehicle, where the tires and camber effects are represented in an extended version of the basic Pacejka Magic Formula. The proposed single-track model is identified and validated using multi-body motorsport libraries which allow simulating the vehicle dynamics properly, especially useful when real experimental data are missing. The fundamental regularization terms and constraints of the controller are tuned to reduce the rate of change of the inputs while assuring an acceptable velocity and path tracking. The motion planning strategy consists of a Fren\'et-Frame-based planner which considers a forecast of the opponent produced by a Kalman filter. The planner chooses the collision-free path and velocity profile to be tracked on a 3 seconds horizon to realize different goals such as following and overtaking. The proposed solution has been applied on a Dallara AV-21 racecar and tested at oval race tracks achieving lateral accelerations up to 25 $m/s^{2}$.
Tremendous progress in deep learning over the last years has led towards a future with autonomous vehicles on our roads. Nevertheless, the performance of their perception systems is strongly dependent on the quality of the utilized training data. As these usually only cover a fraction of all object classes an autonomous driving system will face, such systems struggle with handling the unexpected. In order to safely operate on public roads, the identification of objects from unknown classes remains a crucial task. In this paper, we propose a novel pipeline to detect unknown objects. Instead of focusing on a single sensor modality, we make use of lidar and camera data by combining state-of-the art detection models in a sequential manner. We evaluate our approach on the Waymo Open Perception Dataset and point out current research gaps in anomaly detection.
Implicit neural representations have shown compelling results in offline 3D reconstruction and also recently demonstrated the potential for online SLAM systems. However, applying them to autonomous 3D reconstruction, where robots are required to explore a scene and plan a view path for the reconstruction, has not been studied. In this paper, we explore for the first time the possibility of using implicit neural representations for autonomous 3D scene reconstruction by addressing two key challenges: 1) seeking a criterion to measure the quality of the candidate viewpoints for the view planning based on the new representations, and 2) learning the criterion from data that can generalize to different scenes instead of hand-crafting one. For the first challenge, a proxy of Peak Signal-to-Noise Ratio (PSNR) is proposed to quantify a viewpoint quality. The proxy is acquired by treating the color of a spatial point in a scene as a random variable under a Gaussian distribution rather than a deterministic one; the variance of the distribution quantifies the uncertainty of the reconstruction and composes the proxy. For the second challenge, the proxy is optimized jointly with the parameters of an implicit neural network for the scene. With the proposed view quality criterion, we can then apply the new representations to autonomous 3D reconstruction. Our method demonstrates significant improvements on various metrics for the rendered image quality and the geometry quality of the reconstructed 3D models when compared with variants using TSDF or reconstruction without view planning.
Predicting the future states of surrounding traffic participants and planning a safe, smooth, and socially compliant trajectory accordingly is crucial for autonomous vehicles. There are two major issues with the current autonomous driving system: the prediction module is often decoupled from the planning module and the cost function for planning is hard to specify and tune. To tackle these issues, we propose an end-to-end differentiable framework that integrates prediction and planning modules and is able to learn the cost function from data. Specifically, we employ a differentiable nonlinear optimizer as the motion planner, which takes the predicted trajectories of surrounding agents given by the neural network as input and optimizes the trajectory for the autonomous vehicle, thus enabling all operations in the framework to be differentiable including the cost function weights. The proposed framework is trained on a large-scale real-world driving dataset to imitate human driving trajectories in the entire driving scene and validated in both open-loop and closed-loop manners. The open-loop testing results reveal that the proposed method outperforms the baseline methods across a variety of metrics and delivers planning-centric prediction results, allowing the planning module to output close-to-human trajectories. In closed-loop testing, the proposed method shows the ability to handle complex urban driving scenarios and robustness against the distributional shift that imitation learning methods suffer from. Importantly, we find that joint training of planning and prediction modules achieves better performance than planning with a separate trained prediction module in both open-loop and closed-loop tests. Moreover, the ablation study indicates that the learnable components in the framework are essential to ensure planning stability and performance.
Temporal action segmentation (TAS) aims to classify and locate actions in the long untrimmed action sequence. With the success of deep learning, many deep models for action segmentation have emerged. However, few-shot TAS is still a challenging problem. This study proposes an efficient framework for the few-shot skeleton-based TAS, including a data augmentation method and an improved model. The data augmentation approach based on motion interpolation is presented here to solve the problem of insufficient data, and can increase the number of samples significantly by synthesizing action sequences. Besides, we concatenate a Connectionist Temporal Classification (CTC) layer with a network designed for skeleton-based TAS to obtain an optimized model. Leveraging CTC can enhance the temporal alignment between prediction and ground truth and further improve the segment-wise metrics of segmentation results. Extensive experiments on both public and self-constructed datasets, including two small-scale datasets and one large-scale dataset, show the effectiveness of two proposed methods in improving the performance of the few-shot skeleton-based TAS task.
Training neural networks to perform 3D object detection for autonomous driving requires a large amount of diverse annotated data. However, obtaining training data with sufficient quality and quantity is expensive and sometimes impossible due to human and sensor constraints. Therefore, a novel solution is needed for extending current training methods to overcome this limitation and enable accurate 3D object detection. Our solution for the above-mentioned problem combines semi-pseudo-labeling and novel 3D augmentations. For demonstrating the applicability of the proposed method, we have designed a convolutional neural network for 3D object detection which can significantly increase the detection range in comparison with the training data distribution.
This paper presents recent progress on integrating speech separation and enhancement (SSE) into the ESPnet toolkit. Compared with the previous ESPnet-SE work, numerous features have been added, including recent state-of-the-art speech enhancement models with their respective training and evaluation recipes. Importantly, a new interface has been designed to flexibly combine speech enhancement front-ends with other tasks, including automatic speech recognition (ASR), speech translation (ST), and spoken language understanding (SLU). To showcase such integration, we performed experiments on carefully designed synthetic datasets for noisy-reverberant multi-channel ST and SLU tasks, which can be used as benchmark corpora for future research. In addition to these new tasks, we also use CHiME-4 and WSJ0-2Mix to benchmark multi- and single-channel SE approaches. Results show that the integration of SE front-ends with back-end tasks is a promising research direction even for tasks besides ASR, especially in the multi-channel scenario. The code is available online at //github.com/ESPnet/ESPnet. The multi-channel ST and SLU datasets, which are another contribution of this work, are released on HuggingFace.
The existence of representative datasets is a prerequisite of many successful artificial intelligence and machine learning models. However, the subsequent application of these models often involves scenarios that are inadequately represented in the data used for training. The reasons for this are manifold and range from time and cost constraints to ethical considerations. As a consequence, the reliable use of these models, especially in safety-critical applications, is a huge challenge. Leveraging additional, already existing sources of knowledge is key to overcome the limitations of purely data-driven approaches, and eventually to increase the generalization capability of these models. Furthermore, predictions that conform with knowledge are crucial for making trustworthy and safe decisions even in underrepresented scenarios. This work provides an overview of existing techniques and methods in the literature that combine data-based models with existing knowledge. The identified approaches are structured according to the categories integration, extraction and conformity. Special attention is given to applications in the field of autonomous driving.
This manuscript portrays optimization as a process. In many practical applications the environment is so complex that it is infeasible to lay out a comprehensive theoretical model and use classical algorithmic theory and mathematical optimization. It is necessary as well as beneficial to take a robust approach, by applying an optimization method that learns as one goes along, learning from experience as more aspects of the problem are observed. This view of optimization as a process has become prominent in varied fields and has led to some spectacular success in modeling and systems that are now part of our daily lives.
Breakthroughs in machine learning in the last decade have led to `digital intelligence', i.e. machine learning models capable of learning from vast amounts of labeled data to perform several digital tasks such as speech recognition, face recognition, machine translation and so on. The goal of this thesis is to make progress towards designing algorithms capable of `physical intelligence', i.e. building intelligent autonomous navigation agents capable of learning to perform complex navigation tasks in the physical world involving visual perception, natural language understanding, reasoning, planning, and sequential decision making. Despite several advances in classical navigation methods in the last few decades, current navigation agents struggle at long-term semantic navigation tasks. In the first part of the thesis, we discuss our work on short-term navigation using end-to-end reinforcement learning to tackle challenges such as obstacle avoidance, semantic perception, language grounding, and reasoning. In the second part, we present a new class of navigation methods based on modular learning and structured explicit map representations, which leverage the strengths of both classical and end-to-end learning methods, to tackle long-term navigation tasks. We show that these methods are able to effectively tackle challenges such as localization, mapping, long-term planning, exploration and learning semantic priors. These modular learning methods are capable of long-term spatial and semantic understanding and achieve state-of-the-art results on various navigation tasks.
注冊地址： 北京市海淀區羊坊店路18號2幢3層301-191