自主驾驶的典型轨迹计划通常依赖于预测周围障碍的未来行为。近年来,由于其令人印象深刻的性能,基于深度学习的预测模型已被广泛使用。但是,最近的研究表明,在长尾驾驶场景分布之后,在数据集上训练的深度学习模型将遭受“尾巴”的大量预测错误,这可能会导致计划者的失败。为此,这项工作定义了预测模型不确定性的概念,以量化由于数据稀疏而导致的高错误。此外,这项工作提出了一个轨迹规划师,以考虑对更安全性能的这种预测不确定性。首先,由于培训数据不足而导致的预测模型的不确定性是由集成网络结构估算的。然后,轨迹规划师的设计目的是考虑预测不确定性引起的最坏情况。结果表明,在数据不足引起的预测不确定性下,提出的方法可以提高轨迹计划的安全性。同时,使用足够的数据,该框架不会导致过度保守的结果。这项技术有助于在现实世界的长尾数据分布下提高自动驾驶汽车的安全性和可靠性。
translated by 谷歌翻译
Making safe and human-like decisions is an essential capability of autonomous driving systems and learning-based behavior planning is a promising pathway toward this objective. Distinguished from existing learning-based methods that directly output decisions, this work introduces a predictive behavior planning framework that learns to predict and evaluate from human driving data. Concretely, a behavior generation module first produces a diverse set of candidate behaviors in the form of trajectory proposals. Then the proposed conditional motion prediction network is employed to forecast other agents' future trajectories conditioned on each trajectory proposal. Given the candidate plans and associated prediction results, we learn a scoring module to evaluate the plans using maximum entropy inverse reinforcement learning (IRL). We conduct comprehensive experiments to validate the proposed framework on a large-scale real-world urban driving dataset. The results reveal that the conditional prediction model is able to forecast multiple possible future trajectories given a candidate behavior and the prediction results are reactive to different plans. Moreover, the IRL-based scoring module can properly evaluate the trajectory proposals and select close-to-human ones. The proposed framework outperforms other baseline methods in terms of similarity to human driving trajectories. Moreover, we find that the conditional prediction model can improve both prediction and planning performance compared to the non-conditional model, and learning the scoring module is critical to correctly evaluating the candidate plans to align with human drivers.
translated by 谷歌翻译
相应地预测周围交通参与者的未来状态,并计划安全,平稳且符合社会的轨迹对于自动驾驶汽车至关重要。当前的自主驾驶系统有两个主要问题:预测模块通常与计划模块解耦,并且计划的成本功能很难指定和调整。为了解决这些问题,我们提出了一个端到端的可区分框架,该框架集成了预测和计划模块,并能够从数据中学习成本函数。具体而言,我们采用可区分的非线性优化器作为运动计划者,该运动计划将神经网络给出的周围剂的预测轨迹作为输入,并优化了自动驾驶汽车的轨迹,从而使框架中的所有操作都可以在框架中具有可观的成本,包括成本功能权重。提出的框架经过大规模的现实驾驶数据集进行了训练,以模仿整个驾驶场景中的人类驾驶轨迹,并在开环和闭环界面中进行了验证。开环测试结果表明,所提出的方法的表现优于各种指标的基线方法,并提供以计划为中心的预测结果,从而使计划模块能够输出接近人类的轨迹。在闭环测试中,提出的方法表明能够处理复杂的城市驾驶场景和鲁棒性,以抵抗模仿学习方法所遭受的分配转移。重要的是,我们发现计划和预测模块的联合培训比在开环和闭环测试中使用单独的训练有素的预测模块进行计划要比计划更好。此外,消融研究表明,框架中的可学习组件对于确保计划稳定性和性能至关重要。
translated by 谷歌翻译
基于神经网络的驾驶规划师在改善自动驾驶的任务绩效方面表现出了巨大的承诺。但是,确保具有基于神经网络的组件的系统的安全性,尤其是在密集且高度交互式的交通环境中,这是至关重要的,但又具有挑战性。在这项工作中,我们为基于神经网络的车道更改提出了一个安全驱动的互动计划框架。为了防止过度保守计划,我们确定周围车辆的驾驶行为并评估其侵略性,然后以互动方式相应地适应了计划的轨迹。如果在预测的最坏情况下,即使存在安全的逃避轨迹,则自我车辆可以继续改变车道;否则,它可以停留在当前的横向位置附近或返回原始车道。我们通过广泛而全面的实验环境以及在自动驾驶汽车公司收集的现实情况下进行了广泛的模拟,定量证明了计划者设计的有效性及其优于基线方法的优势。
translated by 谷歌翻译
Accurately predicting interactive road agents' future trajectories and planning a socially compliant and human-like trajectory accordingly are important for autonomous vehicles. In this paper, we propose a planning-centric prediction neural network, which takes surrounding agents' historical states and map context information as input, and outputs the joint multi-modal prediction trajectories for surrounding agents, as well as a sequence of control commands for the ego vehicle by imitation learning. An agent-agent interaction module along the time axis is proposed in our network architecture to better comprehend the relationship among all the other intelligent agents on the road. To incorporate the map's topological information, a Dynamic Graph Convolutional Neural Network (DGCNN) is employed to process the road network topology. Besides, the whole architecture can serve as a backbone for the Differentiable Integrated motion Prediction with Planning (DIPP) method by providing accurate prediction results and initial planning commands. Experiments are conducted on real-world datasets to demonstrate the improvements made by our proposed method in both planning and prediction accuracy compared to the previous state-of-the-art methods.
translated by 谷歌翻译
自治车辆的评估和改善规划需要可扩展的长尾交通方案。有用的是,这些情景必须是现实的和挑战性的,但不能安全地开车。在这项工作中,我们介绍努力,一种自动生成具有挑战性的场景的方法,导致给定的计划者产生不良行为,如冲突。为了维护情景合理性,关键的想法是利用基于图形的条件VAE的形式利用学习的交通运动模型。方案生成在该流量模型的潜在空间中制定了优化,通过扰乱初始的真实世界的场景来产生与给定计划者碰撞的轨迹。随后的优化用于找到“解决方案”的场景,确保改进给定的计划者是有用的。进一步的分析基于碰撞类型的群集生成的场景。我们攻击两名策划者并展示争取在这两种情况下成功地产生了现实,具有挑战性的情景。我们另外“关闭循环”并使用这些方案优化基于规则的策划器的超参数。
translated by 谷歌翻译
自主驾驶的运动预测领域的先前艺术倾向于寻找接近地面真理轨迹的轨迹。但是,这种问题的表述和方法经常导致多样性和偏见轨迹预测的丧失。因此,它们不适合现实世界的自主驾驶,在这种驾驶中,多样化和依赖道路的多模式轨迹预测对安全至关重要。为此,本研究提出了一种新颖的损失函数\ textit {lane损失},可确保地图自适应多样性并适应几何约束。对带有新型轨迹候选建议模块的两阶段轨迹预测架构,\ textit {轨迹预测注意(TPA)}经过训练,通过车道损失训练,鼓励多个轨迹分布多样,以涵盖可行的方式以图像意识的方式涵盖可行的操作。此外,考虑到现有的轨迹性能指标正在重点是基于地面真理未来轨迹评估准确性,因此还建议定量评估指标来评估预测的多个轨迹的多样性。在Argoverse数据集上进行的实验表明,所提出的方法显着提高了预测轨迹的多样性,而无需牺牲预测准确性。
translated by 谷歌翻译
应用强化学习来自动驾驶需要某些挑战,这主要是由于大规模的交通流动,这种挑战是动态变化的。为了应对此类挑战,有必要快速确定对周围车辆不断变化的意图的响应策略。因此,我们提出了一种新的政策优化方法,用于使用基于图的互动感知约束来安全驾驶。在此框架中,运动预测和控制模块是同时训练的,同时共享包含社会环境的潜在表示。此外,为了反映社交互动,我们以图形形式表达了代理的运动并过滤特征。这有助于保留相邻节点的时空位置。此外,我们创建反馈循环以有效地组合这两个模块。结果,这种方法鼓励博学的控制器免受动态风险的侵害,并在各种情况下使运动预测强大。在实验中,我们与城市驾驶模拟器Carla建立了一个包括各种情况的导航场景。该实验表明,与基线相比,导航策略和运动预测的两侧的最新性能。
translated by 谷歌翻译
Autonomous vehicle (AV) stacks are typically built in a modular fashion, with explicit components performing detection, tracking, prediction, planning, control, etc. While modularity improves reusability, interpretability, and generalizability, it also suffers from compounding errors, information bottlenecks, and integration challenges. To overcome these challenges, a prominent approach is to convert the AV stack into an end-to-end neural network and train it with data. While such approaches have achieved impressive results, they typically lack interpretability and reusability, and they eschew principled analytical components, such as planning and control, in favor of deep neural networks. To enable the joint optimization of AV stacks while retaining modularity, we present DiffStack, a differentiable and modular stack for prediction, planning, and control. Crucially, our model-based planning and control algorithms leverage recent advancements in differentiable optimization to produce gradients, enabling optimization of upstream components, such as prediction, via backpropagation through planning and control. Our results on the nuScenes dataset indicate that end-to-end training with DiffStack yields substantial improvements in open-loop and closed-loop planning metrics by, e.g., learning to make fewer prediction errors that would affect planning. Beyond these immediate benefits, DiffStack opens up new opportunities for fully data-driven yet modular and interpretable AV architectures. Project website: https://sites.google.com/view/diffstack
translated by 谷歌翻译
The goal of autonomous vehicles is to navigate public roads safely and comfortably. To enforce safety, traditional planning approaches rely on handcrafted rules to generate trajectories. Machine learning-based systems, on the other hand, scale with data and are able to learn more complex behaviors. However, they often ignore that agents and self-driving vehicle trajectory distributions can be leveraged to improve safety. In this paper, we propose modeling a distribution over multiple future trajectories for both the self-driving vehicle and other road agents, using a unified neural network architecture for prediction and planning. During inference, we select the planning trajectory that minimizes a cost taking into account safety and the predicted probabilities. Our approach does not depend on any rule-based planners for trajectory generation or optimization, improves with more training data and is simple to implement. We extensively evaluate our method through a realistic simulator and show that the predicted trajectory distribution corresponds to different driving profiles. We also successfully deploy it on a self-driving vehicle on urban public roads, confirming that it drives safely without compromising comfort. The code for training and testing our model on a public prediction dataset and the video of the road test are available at https://woven.mobi/safepathnet
translated by 谷歌翻译
轨迹预测对于自动驾驶汽车(AV)是必不可少的,以计划正确且安全的驾驶行为。尽管许多先前的作品旨在达到更高的预测准确性,但很少有人研究其方法的对抗性鲁棒性。为了弥合这一差距,我们建议研究数据驱动的轨迹预测系统的对抗性鲁棒性。我们设计了一个基于优化的对抗攻击框架,该框架利用精心设计的可区分动态模型来生成逼真的对抗轨迹。从经验上讲,我们基于最先进的预测模型的对抗性鲁棒性,并表明我们的攻击使通用指标和计划感知指标的预测错误增加了50%以上和37%。我们还表明,我们的攻击可以导致AV在模拟中驶离道路或碰撞到其他车辆中。最后,我们演示了如何使用对抗训练计划来减轻对抗性攻击。
translated by 谷歌翻译
我们解决了由具有不同驱动程序行为的道路代理人填充的密集模拟交通环境中的自我车辆导航问题。由于其异构行为引起的代理人的不可预测性,这种环境中的导航是挑战。我们提出了一种新的仿真技术,包括丰富现有的交通模拟器,其具有与不同程度的侵略性程度相对应的行为丰富的轨迹。我们在驾驶员行为建模算法的帮助下生成这些轨迹。然后,我们使用丰富的模拟器培训深度加强学习(DRL)策略,包括一组高级车辆控制命令,并在测试时间使用此策略来执行密集流量的本地导航。我们的政策隐含地模拟了交通代理商之间的交互,并计算了自助式驾驶员机动,例如超速,超速,编织和突然道路变化的激进驾驶员演习的安全轨迹。我们增强的行为丰富的模拟器可用于生成由对应于不同驱动程序行为和流量密度的轨迹组成的数据集,我们的行为的导航方案可以与最先进的导航算法相结合。
translated by 谷歌翻译
仿真是对机器人系统(例如自动驾驶汽车)进行扩展验证和验证的关键。尽管高保真物理和传感器模拟取得了进步,但在模拟道路使用者的现实行为方面仍然存在一个危险的差距。这是因为,与模拟物理和图形不同,设计人类行为的第一个原理模型通常是不可行的。在这项工作中,我们采用了一种数据驱动的方法,并提出了一种可以学会从现实世界驱动日志中产生流量行为的方法。该方法通过将交通仿真问题分解为高级意图推理和低级驾驶行为模仿,通过利用驾驶行为的双层层次结构来实现高样本效率和行为多样性。该方法还结合了一个计划模块,以获得稳定的长马行为。我们从经验上验证了我们的方法,即交通模拟(位)的双层模仿,并具有来自两个大规模驾驶数据集的场景,并表明位表明,在现实主义,多样性和长途稳定性方面可以达到平衡的交通模拟性能。我们还探索了评估行为现实主义的方法,并引入了一套评估指标以进行交通模拟。最后,作为我们的核心贡献的一部分,我们开发和开源一个软件工具,该工具将跨不同驱动数据集的数据格式统一,并将现有数据集将场景转换为交互式仿真环境。有关其他信息和视频,请参见https://sites.google.com/view/nvr-bits2022/home
translated by 谷歌翻译
Although extensive research in planning has been carried out for normal scenarios, path planning in emergencies has not been thoroughly explored, especially when vehicles move at a higher speed and have less space for avoiding a collision. For emergency collision avoidance, the controller should have the ability to deal with complicated environments and take collision mitigation into consideration since the problem may have no feasible solution. We propose a safety controller by using model predictive control and artificial potential function. A new artificial potential function inspired by line charge is proposed as the cost function for our model predictive controller. The new artificial potential function takes the shape of all objects into consideration. In particular, the artificial potential function that we proposed has the flexibility to fit the shape of the road structures such as the intersection, while the artificial potential function in most of the previous work could only be used in a highway scenario. Moreover, we could realize collision mitigation for a specific part of the vehicle by increasing the quantity of the charge at the corresponding place. We have tested our methods in 192 cases from 8 different scenarios in simulation. The simulation results show that the success rate of the proposed safety controller is 20% higher than using HJ-reachability with system decomposition. It could also decrease 43% of collision that happens at the pre-assigned part.
translated by 谷歌翻译
到达状态的密度可以帮助理解安全至关重要的系统的风险,尤其是在最坏情况下的情况过于保守的情况下。最近的工作提供了一种数据驱动的方法来计算自主系统在线前进状态的密度分布。在本文中,我们研究了这种方法与模型预测控制在不确定性下的可验证安全路径计划的结合。我们首先使用学习的密度分布来计算在线碰撞的风险。如果这种风险超过可接受的阈值,我们的方法将计划在先前轨迹周围采取新的途径,并在阈值以下碰撞风险。我们的方法非常适合处理具有不确定性和复杂动力学的系统,因为我们的数据驱动方法不需要系统动力学的分析形式,并且可以通过不确定性的任意初始分布来估算正向状态密度。我们设计了两个具有挑战性的场景(自动驾驶和气垫船控制),以在系统不确定性下的障碍物中进行安全运动计划。我们首先表明我们的密度估计方法可以达到与基于蒙特卡洛的方法相似的准确性,同时仅使用0.01倍训练样本。通过利用估计的风险,我们的算法在执行超过0.99的安全速率时达到目标达到最高成功率。
translated by 谷歌翻译
在现代自治堆栈中,预测模块对于在其他移动代理的存在下计划动作至关重要。但是,预测模块的失败会误导下游规划师做出不安全的决定。确实,轨迹预测任务固有的高度不确定性可确保这种错误预测经常发生。由于需要提高自动驾驶汽车的安全而不受损害其性能的需求,我们开发了一个概率运行时监视器,该监视器检测到何时发生“有害”预测故障,即与任务相关的失败检测器。我们通过将轨迹预测错误传播到计划成本来推理其对AV的影响来实现这一目标。此外,我们的检测器还配备了假阳性和假阴性速率的性能度量,并允许进行无数据校准。在我们的实验中,我们将检测器与其他各种检测器进行了比较,发现我们的检测器在接收器操作员特征曲线下具有最高的面积。
translated by 谷歌翻译
在这项工作中,我们提出了世界上第一个基于闭环ML的自动驾驶计划基准。虽然存在基于ML的ML的越来越多的ML的议员,但缺乏已建立的数据集和指标限制了该领域的进展。自主车辆运动预测的现有基准专注于短期运动预测,而不是长期规划。这导致了以前的作品来使用基于L2的度量标准的开放循环评估,这不适合公平地评估长期规划。我们的基准通过引入大规模驾驶数据集,轻量级闭环模拟器和特定于运动规划的指标来克服这些限制。我们提供高质量的数据集,在美国和亚洲的4个城市提供1500h的人类驾驶数据,具有广泛不同的交通模式(波士顿,匹兹堡,拉斯维加斯和新加坡)。我们将提供具有无功代理的闭环仿真框架,并提供一系列一般和方案特定的规划指标。我们计划在Neurips 2021上发布数据集,并在2022年初开始组织基准挑战。
translated by 谷歌翻译
近年来,道路安全引起了智能运输系统领域的研究人员和从业者的重大关注。作为最常见的道路用户群体之一,行人由于其不可预测的行为和运动而导致令人震惊,因为车辆行人互动的微妙误解可以很容易地导致风险的情况或碰撞。现有方法使用预定义的基于碰撞的模型或人类标签方法来估计行人的风险。这些方法通常受到他们的概括能力差,缺乏对自我车辆和行人之间的相互作用的限制。这项工作通过提出行人风险级预测系统来解决所列问题。该系统由三个模块组成。首先,收集车辆角度的行人数据。由于数据包含关于自我车辆和行人的运动的信息,因此可以简化以交互感知方式预测时空特征的预测。使用长短短期存储器模型,行人轨迹预测模块预测后续五个框架中的时空特征。随着预测的轨迹遵循某些交互和风险模式,采用混合聚类和分类方法来探讨时空特征中的风险模式,并使用学习模式训练风险等级分类器。在预测行人的时空特征并识别相应的风险水平时,确定自我车辆和行人之间的风险模式。实验结果验证了PRLP系统的能力,以预测行人的风险程度,从而支持智能车辆的碰撞风险评估,并为车辆和行人提供安全警告。
translated by 谷歌翻译
Modern autonomous driving system is characterized as modular tasks in sequential order, i.e., perception, prediction and planning. As sensors and hardware get improved, there is trending popularity to devise a system that can perform a wide diversity of tasks to fulfill higher-level intelligence. Contemporary approaches resort to either deploying standalone models for individual tasks, or designing a multi-task paradigm with separate heads. These might suffer from accumulative error or negative transfer effect. Instead, we argue that a favorable algorithm framework should be devised and optimized in pursuit of the ultimate goal, i.e. planning of the self-driving-car. Oriented at this goal, we revisit the key components within perception and prediction. We analyze each module and prioritize the tasks hierarchically, such that all these tasks contribute to planning (the goal). To this end, we introduce Unified Autonomous Driving (UniAD), the first comprehensive framework up-to-date that incorporates full-stack driving tasks in one network. It is exquisitely devised to leverage advantages of each module, and provide complementary feature abstractions for agent interaction from a global perspective. Tasks are communicated with unified query design to facilitate each other toward planning. We instantiate UniAD on the challenging nuScenes benchmark. With extensive ablations, the effectiveness of using such a philosophy is proven to surpass previous state-of-the-arts by a large margin in all aspects. The full suite of codebase and models would be available to facilitate future research in the community.
translated by 谷歌翻译
人类行为的不确定性对拥挤的城市环境中的自动驾驶构成了重大挑战。部分可观察到的马尔可夫决策过程(POMDP)为不确定性下的计划提供了一个原则的框架,通常利用蒙特卡洛抽样来实现在线绩效进行复杂的任务。但是,抽样还通过潜在缺失关键事件引起了安全问题。为了解决这个问题,我们提出了一种新的算法,学习对驾驶行为(领导者)的关注,这些算法在计划过程中学习了批判性人类行为。领导者学习了一个神经网络生成器,以实时情况下对人类行为的关注。它将注意力集成到信仰空间计划者中,使用重要性抽样来偏向关键事件。为了训练该算法,我们让注意力生成器和计划者组成了最小游戏。通过解决Min-Max游戏,领导者学会了无需人类标签即可执行风险意识的计划。
translated by 谷歌翻译