Light Detection and Ranging (LIDAR) sensors play an important role in the perception stack of autonomous robots, supplying mapping and localization pipelines with depth measurements of the environment. While their accuracy outperforms other types of depth sensors, such as stereo or time-of-flight cameras, the accurate modeling of LIDAR sensors requires laborious manual calibration that typically does not take into account the interaction of laser light with different surface types, incidence angles and other phenomena that significantly influence measurements. In this work, we introduce a physically plausible model of a 2D continuous-wave LIDAR that accounts for the surface-light interactions and simulates the measurement process in the Hokuyo URG-04LX LIDAR. Through automatic differentiation, we employ gradient-based optimization to estimate model parameters from real sensor measurements.
We propose a centralized control framework to select suitable robots from a heterogeneous pool and place them at appropriate locations to monitor a region for events of interest. In the event of a robot failure, the framework repositions robots in a user-defined local neighborhood of the failed robot to compensate for the coverage loss. The central controller augments the team with additional robots from the robot pool when simply repositioning robots fails to attain a user-specified level of desired coverage. The size of the local neighborhood around the failed robot and the desired coverage over the region are two settings that can be manipulated to achieve a user-specified balance. We investigate the trade-off between the coverage compensation achieved through local repositioning and the computation required to plan the new robot locations. We also study the relationship between the size of the local neighborhood and the number of additional robots added to the team for a given user-specified level of desired coverage. The computational complexity of our resilient strategy (tunable resilient coordination), is quadratic in both neighborhood size and number of robots in the team. At first glance, it seems that any desired level of coverage can be efficiently achieved by augmenting the robot team with more robots while keeping the neighborhood size fixed. However, we show that to reach a high level of coverage in a neighborhood with a large robot population, it is more efficient to enlarge the neighborhood size, instead of adding additional robots and repositioning them.
We address the problem of maintaining resource availability in a networked multi-robot system performing distributed target tracking. In our model, robots are equipped with sensing and computational resources enabling them to track a target's position using a Distributed Kalman Filter (DKF). We use the trace of each robot's sensor measurement noise covariance matrix as a measure of sensing quality. When a robot's sensing quality deteriorates, the system's communication graph is modified by adding edges such that the robot with deteriorating sensor quality may share information with other robots to improve the team's target tracking ability. This computation is performed centrally and is designed to work without a large change in the number of active communication links. We propose two mixed integer semi-definite programming formulations (an 'agent-centric' strategy and a 'team-centric' strategy) to achieve this goal. We implement both formulations and a greedy strategy in simulation and show that the team-centric strategy outperforms the agent-centric and greedy strategies.
Intelligent agents need a physical understanding of the world to predict the impact of their actions in the future. While learning-based models of the environment dynamics have contributed to significant improvements in sample efficiency compared to model-free reinforcement learning algorithms, they typically fail to generalize to system states beyond the training data, while often grounding their predictions on non-interpretable latent variables. We introduce Interactive Differentiable Simulation (IDS), a differentiable physics engine, that allows for efficient, accurate inference of physical properties of rigid-body systems. Integrated into deep learning architectures, our model is able to accomplish system identification using visual input, leading to an interpretable model of the world whose parameters have physical meaning. We present experiments showing automatic task-based robot design and parameter estimation for nonlinear dynamical systems by automatically calculating gradients in IDS. When integrated into an adaptive model-predictive control algorithm, our approach exhibits orders of magnitude improvements in sample efficiency over model-free reinforcement learning algorithms on challenging nonlinear control domains.
Quadrotor stabilizing controllers often require careful, model-specific tuning for safe operation. We use reinforcement learning to train policies in simulation that transfer remarkably well to multiple different physical quadrotors. Our policies are low-level, i.e., we map the rotorcrafts' state directly to the motor outputs. The trained control policies are very robust to external disturbances and can withstand harsh initial conditions such as throws. We show how different training methodologies (change of the cost function, modeling of noise, use of domain randomization) might affect flight performance. To the best of our knowledge, this is the first work that demonstrates that a simple neural network can learn a robust stabilizing low-level quadrotor controller (without the use of a stabilizing PD controller) that is shown to generalize to multiple quadrotors.
Personal robots assisting humans must perform complex manipulation tasks that are typically difficult to specify in traditional motion planning pipelines, where multiple objectives must be met and the high-level context be taken into consideration. Learning from demonstration (LfD) provides a promising way to learn these kind of complex manipulation skills even from non-technical users. However, it is challenging for existing LfD methods to efficiently learn skills that can generalize to task specifications that are not covered by demonstrations. In this paper, we introduce a state transition model (STM) that generates joint-space trajectories by imitating motions from expert behavior. Given a few demonstrations, we show in real robot experiments that the learned STM can quickly generalize to unseen tasks and synthesize motions having longer time horizons than the expert trajectories. Compared to conventional motion planners, our approach enables the robot to accomplish complex behaviors from high-level instructions without laborious hand-engineering of planning objectives, while being able to adapt to changing goals during the skill execution. In conjunction with a trajectory optimizer, our STM can construct a high-quality skeleton of a trajectory that can be further improved in smoothness and precision. In combination with a learned inverse dynamics model, we additionally present results where the STM is used as a high-level planner. A video of our experiments is available at https://youtu.be/85DX9Ojq-90
We propose a method to maintain high resource in a networked heterogeneous multi-robot system to resource failures. In our model, resources such as and computation are available on robots. The robots engaged in a joint task using these pooled resources. In our model, a resource on a particular robot becomes unavailable e.g., a sensor ceases to function due to a failure), the system reconfigures so that the robot continues to have to this resource by communicating with other robots. Specifically, we consider the problem of selecting edges to be in the system's communication graph after a resource has occurred. We define a metric that allows us to characterize the quality of the resource distribution in the represented by the communication graph. Upon a resource becoming unavailable due to failure, we reconfigure network so that the resource distribution is brought as to the ideal resource distribution as possible without a big change in the communication cost. Our approach uses integer semi-definite programming to achieve this goal. We also provide a simulated annealing method to compute a formation that satisfies the inter-robot distances imposed by the topology, along with other constraints. Our method can compute a communication topology, spatial formation, and formation change motion planning in a few seconds. We validate our method in simulation and real-robot experiments with a team of seven quadrotors.
Autonomous underwater vehicles (AUVs) are robotic platforms that are commonly used to map the sea floor, for example for benthic surveys or for naval mine countermeasures (MCM) operations. AUVs create an acoustic image of the survey area, such that objects on the seabed can be identified and, in the case of MCM, mines can be found and disposed of. The common method for creating such seabed maps is to run a lawnmower survey, which is a standard method in coverage path planning. We are interested in exploring alternate techniques for surveying areas of interest, in order to reduce mission time or assess feasible actions, such as finding a safe path through a hazardous region. In this paper, we use Gaussian Process regression to build models of seabed complexity data, obtained through lawnmower surveys. We evaluate several commonly used kernels to assess their modeling performance, which includes modeling discontinuities in the data. Our results show that an additive Mat\'ern kernel is most suitable for modeling seabed complexity data. On top of the GP model, we use adaptations of two standard path planning methods, A* and RRT*, to find safe paths for marine vessels through the modeled areas. We evaluate the planned paths and also run a vehicle dynamics simulator to assess potential performance by a marine vessel.
Recent advancements in machine learning research have given rise to recurrent neural networks that are able to synthesize high-dimensional motion sequences over long time horizons. By leveraging these sequence learning techniques, we introduce a state transition model (STM) that is able to learn a variety of complex motion sequences in joint position space. Given few demonstrations from a motion planner, we show in real robot experiments that the learned STM can quickly generalize to unseen tasks. Our approach enables the robot to accomplish complex behaviors from high-level instructions that would require laborious hand-engineered sequencing of trajectories with traditional motion planners. A video of our experiments is available at https://youtu.be/TRlq_N23XTM
Robots such as autonomous underwater vehicles (AUVs) and autonomous surface vehicles (ASVs) have been used for sensing and monitoring aquatic environments such as oceans and lakes. Environmental sampling is a challenging task because the environmental attributes to be observed can vary both spatially and temporally, and the target environment is usually a large and continuous domain whereas the sampling data is typically sparse and limited. The challenges require that the sampling method must be informative and efficient enough to catch up with the environmental dynamics. In this paper we present a planning and learning method that enables a sampling robot to perform persistent monitoring tasks by learning and refining a dynamic "data map" that models a spatiotemporal environment attribute such as ocean salinity content. Our environmental sampling framework consists of two components: to maximize the information collected, we propose an informative planning component that efficiently generates sampling waypoints that contain the maximal information; To alleviate the computational bottleneck caused by large-scale data accumulated, we develop a component based on a sparse Gaussian Process whose hyperparameters are learned online by taking advantage of only a subset of data that provides the greatest contribution. We validate our method with both simulations running on real ocean data and field trials with an ASV in a lake environment. Our experiments show that the proposed framework is both accurate in learning the environmental data map and efficient in catching up with the dynamic environmental changes.