Posted Content•
Learning Inverse Kinodynamics for Accurate High-Speed Off-Road Navigation on Unstructured Terrain
TL;DR: To capture the complex kinodynamic model and mathematically unknown world state, a kinodynamic planner is learned in a data-driven manner with onboard inertial observations, which enables fast and accurate off-road navigation, and outperforms environment-independent alternatives.
Abstract: This paper presents a learning-based approach to consider the effect of unobservable world states in kinodynamic motion planning in order to enable accurate high-speed off-road navigation on unstructured terrain. Existing kinodynamic motion planners either operate in structured and homogeneous environments and thus do not need to explicitly account for terrain-vehicle interaction, or assume a set of discrete terrain classes. However, when operating on unstructured terrain, especially at high speeds, even small variations in the environment will be magnified and cause inaccurate plan execution. In this paper, to capture the complex kinodynamic model and mathematically unknown world state, we learn a kinodynamic planner in a data-driven manner with onboard inertial observations. Our approach is tested on a physical robot in different indoor and outdoor environments, enables fast and accurate off-road navigation, and outperforms environment-independent alternatives, demonstrating 52.4% to 86.9% improvement in terms of plan execution success rate while traveling at high speeds.
Citations
More filters
TL;DR: This survey presents a comprehensive overview of the current autonomous racing platforms, emphasizing the software-hardware co-evolution to the current stage and a summary of open research challenges that will guide future researchers in this field.
Abstract: The rising popularity of self-driving cars has led to the emergence of a new research field in recent years: Autonomous racing. Researchers are developing software and hardware for high-performance race vehicles which aim to operate autonomously on the edge of the vehicle’s limits: High speeds, high accelerations, low reaction times, highly uncertain, dynamic, and adversarial environments. This paper represents the first holistic survey that covers the research in the field of autonomous racing. We focus on the field of autonomous racecars only and display the algorithms, methods, and approaches used in the areas of perception, planning, control, and end-to-end learning. Further, with an increasing number of autonomous racing competitions, researchers now have access to high-performance platforms to test and evaluate their autonomy algorithms. This survey presents a comprehensive overview of the current autonomous racing platforms, emphasizing the software-hardware co-evolution to the current stage. Finally, based on additional discussion with leading researchers in the field, we conclude with a summary of open research challenges that will guide future researchers in this field.
69 citations
22 Sep 2022
TL;DR: A new class of implicit control policies combining the benefits of imitation learning with the robust handling of system constraints from Model Predictive Control (MPC) are presented, called Performer-MPC, which achieves > 40% better goal reached in cluttered environments and > 65% better on social metrics when navigating around humans.
Abstract: Despite decades of research, existing navigation systems still face real-world challenges when deployed in the wild, e.g., in cluttered home environments or in human-occupied public spaces. To address this, we present a new class of implicit control policies combining the benefits of imitation learning with the robust handling of system constraints from Model Predictive Control (MPC). Our approach, called Performer-MPC, uses a learned cost function parameterized by vision context embeddings provided by Performers -- a low-rank implicit-attention Transformer. We jointly train the cost function and construct the controller relying on it, effectively solving end-to-end the corresponding bi-level optimization problem. We show that the resulting policy improves standard MPC performance by leveraging a few expert demonstrations of the desired navigation behavior in different challenging real-world scenarios. Compared with a standard MPC policy, Performer-MPC achieves>40% better goal reached in cluttered environments and>65% better on social metrics when navigating around humans.
16 citations
Posted Content•
TL;DR: This work proposes a new LfH paradigm that does not require runtime hallucination and can therefore adapt to more realistic navigation scenarios and is tested in a benchmark testbed of 300 simulated navigation environments with a wide range of difficulty levels, and in the real-world.
Abstract: Learning from Hallucination (LfH) is a recent machine learning paradigm for autonomous navigation, which uses training data collected in completely safe environments and adds numerous imaginary obstacles to make the environment densely constrained, to learn navigation planners that produce feasible navigation even in highly constrained (more dangerous) spaces. However, LfH requires hallucinating the robot perception during deployment to match with the hallucinated training data, which creates a need for sometimes-infeasible prior knowledge and tends to generate very conservative planning. In this work, we propose a new LfH paradigm that does not require runtime hallucination -- a feature we call "sober deployment" -- and can therefore adapt to more realistic navigation scenarios. This novel Hallucinated Learning and Sober Deployment (HLSD) paradigm is tested in a benchmark testbed of 300 simulated navigation environments with a wide range of difficulty levels, and in the real-world. In most cases, HLSD outperforms both the original LfH method and a classical navigation planner.
12 citations
Cites background from "Learning Inverse Kinodynamics for A..."
...with obstacle avoidance [1], enabling terrain-based navigation [4], [14], allowing robots to move around humans [2], and tuning parameters for classical navigation systems [5], [15], [16]....
[...]
30 Mar 2022
TL;DR: Visual-Inertial Inverse Kinodynamics is introduced, a novel learning based IKD model that is conditioned on visual information from a terrain patch ahead of the robot in addition to past inertial information, enabling it to anticipate kinodynamic interactions in the future.
Abstract: One of the key challenges in high-speed off-road navigation on ground vehicles is that the kinodynamics of the vehicle-terrain interaction can differ dramatically depending on the terrain. Previous approaches to addressing this challenge have considered learning an inverse kinodynamics (IKD) model, conditioned on inertial information of the vehicle to sense the kinodynamic interactions. In this paper, we hypothesize that to enable accurate high-speed off-road navigation using a learned IKD model, in addition to inertial information from the past, one must also anticipate the kinodynamic interactions of the vehicle with the terrain in the future. To this end, we introduce Visual-Inertial Inverse Kinodynamics (VI-IKD), a novel learning based IKD model that is conditioned on visual information from a terrain patch ahead of the robot in addition to past inertial information, enabling it to anticipate kinodynamic interactions in the future. We validate the effectiveness of VI-IKD in accurate high-speed off-road navigation experimentally on a scale 1/5 UT-AlphaTruck off-road autonomous vehicle in both indoor and outdoor environments and show that compared to other state-of-the-art approaches, VI-IKD enables more accurate and robust off-road navigation on a variety of different terrains at speeds of up to 3.5m/s.
12 citations
TL;DR: In this paper , a receding-horizon reinforcement learning approach for kinodynamic motion planning (RHRL-KDP) of autonomous vehicles in the presence of inaccurate dynamics information and moving obstacles is presented.
Abstract: Kinodynamic motion planning is critical for autonomous vehicles with high maneuverability in dynamic environments. However, obtaining near-optimal motion planning solutions with low computational costs and inaccurate prior model information is challenging. To address this issue, this paper proposes a receding-horizon reinforcement learning approach for kinodynamic motion planning (RHRL-KDP) of autonomous vehicles in the presence of inaccurate dynamics information and moving obstacles. Specifically, a receding-horizon actor-critic reinforcement learning algorithm is presented, resulting in a neural network-based planning strategy that can be learned both offline and online. A neural network-based model is built and learned online to approximate the modeling uncertainty of the prior nominal model in order to improve planning performance. Furthermore, active collision avoidance in dynamic environments is realized by constructing safety-related terms in actor and critic networks using potential fields. In theory, the uniformly ultimate boundedness property of the modeling uncertainty’s approximation error is proven, and the convergence of the proposed RHRL-KDP is analyzed. Simulation tests show that our approach outperforms the previously developed motion planners based on model predictive control (MPC), safe RL, and RRT${^\star }$ in terms of planning performance. Furthermore, in both online and offline learning scenarios, RHRL-KDP outperforms MPC and RRT$^{\star }$ in terms of computational efficiency.
10 citations
References
More filters
TL;DR: This approach, designed for mobile robots equipped with synchro-drives, is derived directly from the motion dynamics of the robot and safely controlled the mobile robot RHINO in populated and dynamic environments.
Abstract: This approach, designed for mobile robots equipped with synchro-drives, is derived directly from the motion dynamics of the robot. In experiments, the dynamic window approach safely controlled the mobile robot RHINO at speeds of up to 95 cm/sec, in populated and dynamic environments.
2,886 citations
"Learning Inverse Kinodynamics for A..." refers background in this paper
...Note that the baseline represents classical modelbased local planners [2], while the ablation is equivalent to existing learning-based local planners, e....
[...]
...CURRENT mobile robot navigation methods can navigate a robot from one point to another safely and reliably in structured and homogeneous environments [1], [2], such as indoor hallways or outdoor paved surfaces....
[...]
02 May 1993
TL;DR: Elastic bands are proposed as the basis for a framework to close the gap between global path planning and real-time sensor-based robot control, enabling the robot to accommodate uncertainties and react to unexpected and moving obstacles.
Abstract: Elastic bands are proposed as the basis for a framework to close the gap between global path planning and real-time sensor-based robot control. An elastic band is a deformable collision-free path. The initial shape of the elastic is the free path generated by a planner. Subjected to artificial forces, the elastic band deforms in real time to a short and smooth path that maintains clearance from the obstacles. The elastic continues to deform as changes in the environment are detected by sensors, enabling the robot to accommodate uncertainties and react to unexpected and moving obstacles. While providing a tight connection between the robot and its environment, the elastic band preserves the global nature of the planned path. The framework is outlined, and an efficient implementation based on bubbles is discussed. >
818 citations
"Learning Inverse Kinodynamics for A..." refers background in this paper
...CURRENT mobile robot navigation methods can navigate a robot from one point to another safely and reliably in structured and homogeneous environments [1], [2], such as indoor hallways or outdoor paved surfaces....
[...]
17 Dec 2016
TL;DR: This work proposes a different approach to perceive forest trials based on a deep neural network used as a supervised image classifier that outperforms alternatives, and yields an accuracy comparable to the accuracy of humans that are tested on the same image classification task.
Abstract: We study the problem of perceiving forest or mountain trails from a single monocular image acquired from the viewpoint of a robot traveling on the trail itself. Previous literature focused on trail segmentation, and used low-level features such as image saliency or appearance contrast; we propose a different approach based on a deep neural network used as a supervised image classifier. By operating on the whole image at once, our system outputs the main direction of the trail compared to the viewing direction. Qualitative and quantitative results computed on a large real-world dataset (which we provide for download) show that our approach outperforms alternatives, and yields an accuracy comparable to the accuracy of humans that are tested on the same image classification task. Preliminary results on using this information for quadrotor control in unseen trails are reported. To the best of our knowledge, this is the first letter that describes an approach to perceive forest trials, which is demonstrated on a quadrotor micro aerial vehicle.
682 citations
"Learning Inverse Kinodynamics for A..." refers background in this paper
...with a different avenue to investigate those perception problems in off-road navigation, researchers also started to combine perception, planning, and motion control using end-to-end learning in unstructured environments [15], [16]....
[...]
Proceedings Article•
05 Dec 2005TL;DR: A vision-based obstacle avoidance system for off-road mobile robots that is trained from end to end to map raw input images to steering angles and exhibits an excellent ability to detect obstacles and navigate around them in real time at speeds of 2 m/s.
Abstract: We describe a vision-based obstacle avoidance system for off-road mobile robots. The system is trained from end to end to map raw input images to steering angles. It is trained in supervised mode to predict the steering angles provided by a human driver during training runs collected in a wide variety of terrains, weather conditions, lighting conditions, and obstacle types. The robot is a 50cm off-road truck, with two forward-pointing wireless color cameras. A remote computer processes the video and controls the robot via radio. The learning system is a large 6-layer convolutional network whose input is a single left/right pair of unprocessed low-resolution images. The robot exhibits an excellent ability to detect obstacles and navigate around them in real time at speeds of 2 m/s.
538 citations
"Learning Inverse Kinodynamics for A..." refers background in this paper
...with a different avenue to investigate those perception problems in off-road navigation, researchers also started to combine perception, planning, and motion control using end-to-end learning in unstructured environments [15], [16]....
[...]
01 May 2017
TL;DR: In this paper, a target-oriented end-to-end navigation model for a robotic platform is learned from expert demonstrations generated in simulation with an existing motion planner, which can safely navigate the robot through obstacle-cluttered environments to reach the provided targets.
Abstract: Learning from demonstration for motion planning is an ongoing research topic. In this paper we present a model that is able to learn the complex mapping from raw 2D-laser range findings and a target position to the required steering commands for the robot. To our best knowledge, this work presents the first approach that learns a target-oriented end-to-end navigation model for a robotic platform. The supervised model training is based on expert demonstrations generated in simulation with an existing motion planner. We demonstrate that the learned navigation model is directly transferable to previously unseen virtual and, more interestingly, real-world environments. It can safely navigate the robot through obstacle-cluttered environments to reach the provided targets. We present an extensive qualitative and quantitative evaluation of the neural network-based motion planner, and compare it to a grid-based global approach, both in simulation and in real-world experiments.
224 citations