Showing papers on "Motion planning published in 2018"

Planning and Decision-Making for Autonomous Vehicles

Abstract: In this review, we provide an overview of emerging trends and challenges in the field of intelligent and autonomous, or self-driving, vehicles. Recent advances in the field of perception, planning,...

Motion Planning Among Dynamic, Decision-Making Agents with Deep Reinforcement Learning

Abstract: Robots that navigate among pedestrians use collision avoidance algorithms to enable safe and efficient operation. Recent works present deep reinforcement learning as a framework to model the complex interactions and cooperation. However, they are implemented using key assumptions about other agents' behavior that deviate from reality as the number of agents in the environment increases. This work extends our previous approach to develop an algorithm that learns collision avoidance among a variety of types of dynamic agents without assuming they follow any particular behavior rules. This work also introduces a strategy using LSTM that enables the algorithm to use observations of an arbitrary number of other agents, instead of previous methods that have a fixed observation size. The proposed algorithm outperforms our previous approach in simulation as the number of agents increases, and the algorithm is demonstrated on a fully autonomous robotic vehicle traveling at human walking speed.

Learning Sampling Distributions for Robot Motion Planning

Abstract: A defining feature of sampling-based motion planning is the reliance on an implicit representation of the state space, which is enabled by a set of probing samples. Traditionally, these samples are drawn either probabilistically or deterministically to uniformly cover the state space. Yet, the motion of many robotic systems is often restricted to “small” regions of the state space, due to e.g. differential constraints or collision-avoidance constraints. To accelerate the planning process, it is thus desirable to devise non-uniform sampling strategies that favor sampling in those regions where an optimal solution might lie. This paper proposes a methodology for nonuniform sampling, whereby a sampling distribution is learned from demonstrations, and then used to bias sampling. The sampling distribution is computed through a conditional variational autoencoder, allowing sample generation from the latent space conditioned on the specific planning problem. This methodology is general, can be used in combination with any sampling-based planner, and can effectively exploit the underlying structure of a planning problem while maintaining the theoretical guarantees of sampling-based approaches. Specifically, on several planning problems, the proposed methodology is shown to effectively learn representations for the relevant regions of the state space, resulting in an order of magnitude improvement in terms of success rate and convergence to the optimal cost.

Complex-YOLO: An Euler-Region-Proposal for Real-time 3D Object Detection on Point Clouds

Abstract: Lidar based 3D object detection is inevitable for autonomous driving, because it directly links to environmental understanding and therefore builds the base for prediction and motion planning. The capacity of inferencing highly sparse 3D data in real-time is an ill-posed problem for lots of other application areas besides automated vehicles, e.g. augmented reality, personal robotics or industrial automation. We introduce Complex-YOLO, a state of the art real-time 3D object detection network on point clouds only. In this work, we describe a network that expands YOLOv2, a fast 2D standard object detector for RGB images, by a specific complex regression strategy to estimate multi-class 3D boxes in Cartesian space. Thus, we propose a specific Euler-Region-Proposal Network (E-RPN) to estimate the pose of the object by adding an imaginary and a real fraction to the regression network. This ends up in a closed complex space and avoids singularities, which occur by single angle estimations. The E-RPN supports to generalize well during training. Our experiments on the KITTI benchmark suite show that we outperform current leading methods for 3D object detection specifically in terms of efficiency. We achieve state of the art results for cars, pedestrians and cyclists by being more than five times faster than the fastest competitor. Further, our model is capable of estimating all eight KITTI-classes, including Vans, Trucks or sitting pedestrians simultaneously with high accuracy.

PRM-RL: Long-range Robotic Navigation Tasks by Combining Reinforcement Learning and Sampling-Based Planning

Abstract: We present PRM-RL, a hierarchical method for long-range navigation task completion that combines sampling-based path planning with reinforcement learning (RL). The RL agents learn short-range, point-to-point navigation policies that capture robot dynamics and task constraints without knowledge of the large-scale topology. Next, the sampling-based planners provide roadmaps which connect robot configurations that can be successfully navigated by the RL agent. The same RL agents are used to control the robot under the direction of the planning, enabling long-range navigation. We use the Probabilistic Roadmaps (PRMs) for the sampling-based planner. The RL agents are constructed using feature-based and deep neural net policies in continuous state and action spaces. We evaluate PRM-RL, both in simulation and on-robot, on two navigation tasks with non-trivial robot dynamics: end-to-end differential drive indoor navigation in office environments, and aerial cargo delivery in urban environments with load displacement constraints. Our results show improvement in task completion over both RL agents on their own and traditional sampling-based planners. In the indoor navigation task, PRM-RL successfully completes up to 215 m long trajectories under noisy sensor conditions, and the aerial cargo delivery completes flights over 1000 m without violating the task constraints in an environment 63 million times larger than used in training.

Survey on computational-intelligence-based UAV path planning

Abstract: The key objective of unmanned aerial vehicle (UAV) path planning is to produce a flight path that connects a start state and a goal state while meeting the required constraints. Computational intelligence (CI) is a set of nature-inspired computational methodologies and approaches for addressing complex real-world problems for which mathematical or traditional modelling does not perform well. It has been applied in the field of UAVs since it can yield effective, accurate and rapid solutions. This article provides an overview of studies on UAV path planning based on CI methods published in major journals and conference proceedings. We survey relevant studies with respect to different CI algorithms utilized in UAV path planning, the types of time domain in UAV path planning, namely, offline and online, and the types of environment models, namely, 2D and 3D. It is observed that CI methods outperform traditional methods on online and 3D problems. The analysis is useful for identifying key results from UAV path planning research and is leveraged in this article to highlight trends and open issues.

How would surround vehicles move? A Unified Framework for Maneuver Classification and Motion Prediction

Abstract: Reliable prediction of surround vehicle motion is a critical requirement for path planning for autonomous vehicles. In this paper we propose a unified framework for surround vehicle maneuver classification and motion prediction that exploits multiple cues, namely, the estimated motion of vehicles, an understanding of typical motion patterns of freeway traffic and inter-vehicle interaction. We report our results in terms of maneuver classification accuracy and mean and median absolute error of predicted trajectories against the ground truth for real traffic data collected using vehicle mounted sensors on freeways. An ablative analysis is performed to analyze the relative importance of each cue for trajectory prediction. Additionally, an analysis of execution time for the components of the framework is presented. Finally, we present multiple case studies analyzing the outputs of our model for complex traffic scenarios

How Would Surround Vehicles Move? A Unified Framework for Maneuver Classification and Motion Prediction

Abstract: Reliable prediction of surround vehicle motion is a critical requirement for path planning for autonomous vehicles. In this paper, we propose a unified framework for surround vehicle maneuver classification and motion prediction that exploits multiple cues, namely, the estimated motion of vehicles, an understanding of typical motion patterns of freeway traffic and intervehicle interaction. We report our results in terms of maneuver classification accuracy and mean and median absolute error of predicted trajectories against the ground truth for real traffic data collected using vehicle mounted sensors on freeways. An ablative analysis is performed to analyze the relative importance of each cue for trajectory prediction. Additionally, an analysis of execution time for the components of the framework is presented. Finally, we present multiple case studies analyzing the outputs of our model for complex traffic scenarios.

Receding horizon path planning for 3D exploration and surface inspection

Abstract: Within this paper a new path planning algorithm for autonomous robotic exploration and inspection is presented. The proposed method plans online in a receding horizon fashion by sampling possible future configurations in a geometric random tree. The choice of the objective function enables the planning for either the exploration of unknown volume or inspection of a given surface manifold in both known and unknown volume. Application to rotorcraft Micro Aerial Vehicles is presented, although planning for other types of robotic platforms is possible, even in the absence of a boundary value solver and subject to nonholonomic constraints. Furthermore, the method allows the integration of a wide variety of sensor models. The presented analysis of computational complexity and thorough simulations-based evaluation indicate good scaling properties with respect to the scenario complexity. Feasibility and practical applicability are demonstrated in real-life experimental test cases with full on-board computation.

Online Safe Trajectory Generation for Quadrotors Using Fast Marching Method and Bernstein Basis Polynomial

Abstract: In this paper, we propose a framework for online quadrotor motion planning for autonomous navigation in unknown environments. Based on the onboard state estimation and environment perception, we adopt a fast marching-based path searching method to find a path on a velocity field induced by the Euclidean signed distance field (ESDF) of the map, to achieve better time allocation. We generate a flight corridor for the quadrotor to travel through by inflating the path against the environment. We represent the trajectory as piecewise Bezier curves by using Bernstein polynomial basis and formulate the trajectory generation problem as typical convex programs. By using Bezier curves, we are able to bound positions and higher order dynamics of the trajectory entirely within safe regions. The proposed motion planning method is integrated into a customized light-weight quadrotor platform and is validated by presenting fully autonomous navigation in unknown cluttered indoor and outdoor environments. We also release our code for trajectory generation as an open-source package.

Path Planning for the Mobile Robot: A Review

Abstract: Good path planning technology of mobile robot can not only save a lot of time, but also reduce the wear and capital investment of mobile robot. Several methodologies have been proposed and reported in the literature for the path planning of mobile robot. Although these methodologies do not guarantee an optimal solution, they have been successfully applied in their works. The purpose of this paper is to review the modeling, optimization criteria and solution algorithms for the path planning of mobile robot. The survey shows GA (genetic algorithm), PSO (particle swarm optimization algorithm), APF (artificial potential field), and ACO (ant colony optimization algorithm) are the most used approaches to solve the path planning of mobile robot. Finally, future research is discussed which could provide reference for the path planning of mobile robot.

Path Smoothing Techniques in Robot Navigation: State-of-the-Art, Current and Future Challenges

Abstract: Robot navigation is an indispensable component of any mobile service robot. Many path planning algorithms generate a path which has many sharp or angular turns. Such paths are not fit for mobile robot as it has to slow down at these sharp turns. These robots could be carrying delicate, dangerous, or precious items and executing these sharp turns may not be feasible kinematically. On the contrary, smooth trajectories are often desired for robot motion and must be generated while considering the static and dynamic obstacles and other constraints like feasible curvature, robot and lane dimensions, and speed. The aim of this paper is to succinctly summarize and review the path smoothing techniques in robot navigation and discuss the challenges and future trends. Both autonomous mobile robots and autonomous vehicles (outdoor robots or self-driving cars) are discussed. The state-of-the-art algorithms are broadly classified into different categories and each approach is introduced briefly with necessary background, merits, and drawbacks. Finally, the paper discusses the current and future challenges in optimal trajectory generation and smoothing research.

Baidu Apollo EM Motion Planner.

Abstract: In this manuscript, we introduce a real-time motion planning system based on the Baidu Apollo (open source) autonomous driving platform. The developed system aims to address the industrial level-4 motion planning problem while considering safety, comfort and scalability. The system covers multilane and single-lane autonomous driving in a hierarchical manner: (1) The top layer of the system is a multilane strategy that handles lane-change scenarios by comparing lane-level trajectories computed in parallel. (2) Inside the lane-level trajectory generator, it iteratively solves path and speed optimization based on a Frenet frame. (3) For path and speed optimization, a combination of dynamic programming and spline-based quadratic programming is proposed to construct a scalable and easy-to-tune framework to handle traffic rules, obstacle decisions and smoothness simultaneously. The planner is scalable to both highway and lower-speed city driving scenarios. We also demonstrate the algorithm through scenario illustrations and on-road test results. The system described in this manuscript has been deployed to dozens of Baidu Apollo autonomous driving vehicles since Apollo v1.5 was announced in September 2017. As of May 16th, 2018, the system has been tested under 3,380 hours and approximately 68,000 kilometers (42,253 miles) of closed-loop autonomous driving under various urban scenarios. The algorithm described in this manuscript is available at https://github.com/ApolloAuto/apollo/tree/master/modules/planning.

Robust Rough-Terrain Locomotion with a Quadrupedal Robot

Abstract: Robots working in natural, urban, and industrial settings need to be able to navigate challenging environments. In this paper, we present a motion planner for the perceptive rough-terrain locomotion with quadrupedal robots. The planner finds safe footholds along with collision-free swing-leg motions by leveraging an acquired terrain map. To this end, we present a novel pose optimization approach that enables the robot to climb over significant obstacles. We experimentally validate our approach with the quadrupedal robot ANYmal by autonomously traversing obstacles such steps, inclines, and stairs. The locomotion planner re-plans the motion at every step to cope with disturbances and dynamic environments. The robot has no prior knowledge of the scene, and all mapping, state estimation, control, and planning is performed in real-time onboard the robot.

An Efficient Acyclic Contact Planner for Multiped Robots

Abstract: We present a contact planner for complex legged locomotion tasks: standing up, climbing stairs using a handrail, crossing rubble, and getting out of a car. The need for such a planner was shown at the DARPA Robotics Challenge, where such behaviors could not be demonstrated (except for egress). Current planners suffer from their prohibitive algorithmic complexity because they deploy a tree of robot configurations projected in contact with the environment. We tackle this issue by introducing a reduction property: the reachability condition. This condition defines a geometric approximation of the contact manifold, which is of low dimension, presents a Cartesian topology, and can be efficiently sampled and explored. The hard contact planning problem can then be decomposed into two subproblems: first, we plan a path for the root without considering the whole-body configuration, using a sampling-based algorithm; then, we generate a discrete sequence of whole-body configurations in static equilibrium along this path, using a deterministic contact-selection algorithm. The reduction breaks the algorithm complexity encountered in previous works, resulting in the first interactive implementation of a contact planner (open source). While no contact planner has yet been proposed with theoretical completeness, we empirically show the interest of our framework: in a few seconds, with high success rates, we generate complex contact plans for various scenarios and two robots: HRP-2 and HyQ. These plans are validated in dynamic simulations or on the real HRP-2 robot.

Tracking Control of Robot Manipulators with Unknown Models: A Jacobian-Matrix-Adaption Method

Abstract: Tracking control of robot manipulators is a fundamental and significant problem in robotic industry. As a conventional solution, the Jacobian-matrix-pseudo-inverse (JMPI) method suffers from two major limitations: one is the requirement on known information of the robot model such as parameter and structure; the other is the position error accumulation phenomenon caused by the open-loop nature. To overcome such two limitations, this paper proposes a novel Jacobian-matrix-adaption (JMA) method for the tracking control of robot manipulators via the zeroing dynamics. Unlike existing works requiring the information of the known robot model, the proposed JMA method uses only the input-output information to control the robot with unknown model. The solution based on the JMA method transforms the internal, implicit, and unmeasurable model information to the external, explicit, and measurable input-output information. Moreover, simulation studies including comparisons and tests substantiate the efficacy and superiority of the proposed JMA method for the tracking control of robot manipulators subject to unknown models.

Fast Genetic Algorithm Path Planner for Fixed-Wing Military UAV Using GPU

Abstract: Military unmanned aerial vehicles (UAVs) are employed in highly dynamic environments and must often adjust their trajectories based on the evolving situation. To operate autonomously and safely, a UAV must be equipped with a path planning module capable of quickly recalculating a feasible and quasi-optimal path in flight while in the event a new obstacle or threat has been detected or simply if the destination point is changed during the mission. To allow for a fast path planning, this paper proposes a parallel implementation of the genetic algorithm on graphics processing unit (GPU). The trajectories are built as series of line segments connected by circular arcs resulting in smooth paths suitable for fixed-wing UAVs. The fitness function we defined takes into account the dynamic constraints of the UAVs and aims to minimize fuel consumption and average flying altitude in order to improve range and avoid detection by enemy radars. This fitness function is also implemented on the GPU and different parallelization strategies were developed and tested for each step of the fitness evaluation. By exploiting the massively parallel architecture of GPUs, the execution time of the proposed path planner was reduced by a factor of 290x compared to a sequential execution on CPU. The path planning module developed was tested using 18 scenarios on six realistic three-dimensional terrains with multiple no-fly zones. We found that the proposed GPU-based path planner was able to find quasi-optimal solutions in a timely fashion allowing in-flight planning.

A Method on Dynamic Path Planning for Robotic Manipulator Autonomous Obstacle Avoidance Based on an Improved RRT Algorithm.

Abstract: In a future intelligent factory, a robotic manipulator must work efficiently and safely in a Human-Robot collaborative and dynamic unstructured environment. Autonomous path planning is the most important issue which must be resolved first in the process of improving robotic manipulator intelligence. Among the path-planning methods, the Rapidly Exploring Random Tree (RRT) algorithm based on random sampling has been widely applied in dynamic path planning for a high-dimensional robotic manipulator, especially in a complex environment because of its probability completeness, perfect expansion, and fast exploring speed over other planning methods. However, the existing RRT algorithm has a limitation in path planning for a robotic manipulator in a dynamic unstructured environment. Therefore, an autonomous obstacle avoidance dynamic path-planning method for a robotic manipulator based on an improved RRT algorithm, called Smoothly RRT (S-RRT), is proposed. This method that targets a directional node extends and can increase the sampling speed and efficiency of RRT dramatically. A path optimization strategy based on the maximum curvature constraint is presented to generate a smooth and curved continuous executable path for a robotic manipulator. Finally, the correctness, effectiveness, and practicability of the proposed method are demonstrated and validated via a MATLAB static simulation and a Robot Operating System (ROS) dynamic simulation environment as well as a real autonomous obstacle avoidance experiment in a dynamic unstructured environment for a robotic manipulator. The proposed method not only provides great practical engineering significance for a robotic manipulator's obstacle avoidance in an intelligent factory, but also theoretical reference value for other type of robots' path planning.

Know Rob 2.0 — A 2nd Generation Knowledge Processing Framework for Cognition-Enabled Robotic Agents

Abstract: In this paper we present KnowRob2, a second generation knowledge representation and reasoning framework for robotic agents. KnowRob2 is an extension and partial redesign of KnowRob, currently one of the most advanced knowledge processing systems for robots that has enabled them to successfully perform complex manipulation tasks such as making pizza, conducting chemical experiments, and setting tables. The knowledge base appears to be a conventional first-order time interval logic knowledge base, but it exists to a large part only virtually: many logical expressions are constructed on demand from data structures of the control program, computed through robotics algorithms including ones for motion planning and solving inverse kinematics problems, and log data stored in noSQL databases. Novel features and extensions of KnowRob2 substantially increase the capabilities of robotic agents of acquiring open-ended manipulation skills and competence, reasoning about how to perform manipulation actions more realistically, and acquiring commonsense knowledge.

Dynamic Locomotion Through Online Nonlinear Motion Optimization for Quadrupedal Robots

Abstract: This letter presents a realtime motion planning and control method that enables a quadrupedal robot to execute dynamic gaits including trot, pace, and dynamic lateral walk, as well as gaits with full flight phases such as jumping, pronking, and running trot. The proposed method also enables smooth transitions between these gaits. Our approach relies on an online zero-moment point based motion planner which continuously updates the reference motion trajectory as a function of the contact schedule and the state of the robot. The reference footholds for each leg are obtained by solving a separate optimization problem. The resulting optimized motion plans are tracked by a hierarchical whole-body controller. Our framework has been tested in simulation and on ANYmal, a fully torque-controllable quadrupedal robot, both in simulation and on the actual robot.

Multi-objective path planning for unmanned surface vehicle with currents effects.

Abstract: This paper investigates the path planning problem for unmanned surface vehicle (USV), wherein the goal is to find the shortest, smoothest, most economical and safest path in the presence of obstacles and currents, which is subject to the collision avoidance, motion boundaries and velocity constraints. We formulate this problem as a multi-objective nonlinear optimization problem with generous constraints. Then, we propose the dynamic augmented multi-objective particle swarm optimization algorithm to achieve the solution. With our approach, USV can select the ideal path from the Pareto optimal paths set. Numerical simulations verify the effectiveness of our formulated model and proposed algorithm.

Learning Robotic Assembly from CAD

Abstract: In this work, motivated by recent manufacturing trends, we investigate autonomous robotic assembly. Industrial assembly tasks require contact-rich manipulation skills, which are challenging to acquire using classical control and motion planning approaches. Consequently, robot controllers for assembly domains are presently engineered to solve a particular task, and cannot easily handle variations in the product or environment. Reinforcement learning (RL) is a promising approach for autonomously acquiring robot skills that involve contact-rich dynamics. However, RL relies on random exploration for learning a control policy, which requires many robot executions, and often gets trapped in locally suboptimal solutions. Instead, we posit that prior knowledge, when available, can improve RL performance. We exploit the fact that in modern assembly domains, geometric information about the task is readily available via the CAD design files. We propose to leverage this prior knowledge by guiding RL along a geometric motion plan, calculated using the CAD data. We show that our approach effectively improves over traditional control approaches for tracking the motion plan, and can solve assembly tasks that require high precision, even without accurate state estimation. In addition, we propose a neural network architecture that can learn to track the motion plan, thereby generalizing the assembly controller to changes in the object positions.

Deep Reinforcement Learning for Interference-Aware Path Planning of Cellular-Connected UAVs

Abstract: In this paper, an interference-aware path planning scheme for a network of cellular-connected unmanned aerial vehicles (UAVs) is proposed. In particular, each UAV acts as a cellular user equipment (UE) and aims at achieving a tradeoff between maximizing energy efficiency and minimizing both wireless latency and the interference caused on the ground network along its path. The problem is cast as a dynamic game among UAVs. To solve this game, a deep reinforcement learning algorithm, based on echo state network (ESN) cells, is proposed. The introduced deep ESN architecture is trained to allow each UAV to map each observation of the network state to an action, with the goal of minimizing a sequence of time-dependent utility functions. Each UAV uses ESN to learn its optimal path, transmission power, and cell association vector at different locations along its path. The proposed algorithm is shown to reach a subgame perfect Nash equilibrium upon convergence. Simulation results show that the proposed scheme achieves better wireless latency per UAV and rate per ground UE while requiring a number of steps that is comparable to a heuristic baseline that considers moving via the shortest distance towards the corresponding destinations.

Sampling-Based Methods for Motion Planning with Constraints

Abstract: Robots with many degrees of freedom (eg, humanoid robots and mobile manipulators) have increasingly been employed to accomplish realistic tasks in domains such as disaster relief, spacecraft logi

Cooperative Collision Avoidance for Nonholonomic Robots

Abstract: In this paper, we present a method, namely $\epsilon$ CCA, for collision avoidance in dynamic environments among interacting agents, such as other robots or humans. Given a preferred motion by a global planner or driver, the method computes a collision-free local motion for a short time horizon, which respects the actuator constraints and allows for smooth and safe control. The method builds on the concept of reciprocal velocity obstacles and extends it to respect the kinodynamic constraints of the robot and account for a grid-based map representation of the environment. The method is best suited for large multirobot settings, including heterogeneous teams of robots, in which computational complexity is of paramount importance and the robots interact with one another. In particular, we consider a set of motion primitives for the robot and solve an optimization in the space of control velocities with additional constraints. Additionally, we propose a cooperative approach to compute safe velocity partitions in the distributed case. We describe several instances of the method for distributed and centralized operation and formulated both as convex and nonconvex optimizations. We compare the different variants and describe the benefits and tradeoffs both theoretically and in extensive experiments with various robotic platforms: robotic wheelchairs, robotic boats, humanoid robots, small unicycle robots, and simulated cars.