• Corpus ID: 12848641

Planning and Control in Unstructured Terrain

  title={Planning and Control in Unstructured Terrain},
  author={Brian P. Gerkey},
We consider the problem of autonomous navigation in an unstructured outdoor environment. We describe the planning and control aspects of an implemented system that drives a robot at modest speeds (∼1 m/s) over a variety of outdoor terrain. In real time, we use a gradient technique to plan globally optimal paths on a cost map, then employ a predictive dynamic controller to compute local velocity commands. Our planner and controller are considered the “best in class” among 10 teams competing in… 

Figures from this paper

Supervised global path planning for mobile robots with obstacle avoidance

A three-level path planning procedure, which allows the imposition of a set of waypoints and the avoidance of dynamic obstacles appearing during the robot motion, thanks to the action of a local planner, has been experimentally tested to plan the motion of a differential mobile robot.

Dynamic parameter update for robot navigation systems through unsupervised environmental situational analysis

This paper proposes a novel approach to automatically detect the type of surrounding environment based on navigation complexity using unsupervised clustering, and limit the local controller's sampling space.

Parallel Algorithm for Precise Navigation Using Black-Box Forward Model and Motion Primitives

A new planning architecture that allows wheeled vehicles to navigate safely in cluttered environments and is deterministic, computationally efficient, runs in constant time, and can be used on platforms of arbitrary shape and drive type.

Real-time autonomous 3D navigation for tracked vehicles in rescue environments

A novel framework for 3D autonomous navigation for tracked vehicles that takes care of clustering and segmentation of point clouds, traversability analysis, autonomous 3D path planning, motion planning and flippers control is presented.

Local exploration strategies for a mobile robot in a highly dynamic environment

The comparison of the strategies shows that the goal planning strategy performs far better than the other two, warranting a significant increase in the complexity and thus computational resources required for successful implementation onto physical platforms.

A hybrid approach for autonomous navigation of mobile robots in partially-known environments

Path Planning Tolerant to Degraded Locomotion Conditions

  • Xiaoling LongS. Schwertfeger
  • Computer Science, Engineering
    2019 IEEE International Symposium on Safety, Security, and Rescue Robotics (SSRR)
  • 2019
An adaptive path planning algorithm is presented that can automatically detect and adapt to different locomotion conditions and reflect those in the planned paths by constantly observing the executed motions of the robot via simultaneously localization and mapping (SLAM).

Real-time perception-guided motion planning for a personal robot

A modular and distributed architecture is proposed, which seamlessly integrates the creation of 3D maps for collision detection and semantic annotations, with a real-time motion replanning framework.

Real-time Motion Planning Based on Layered Cost Map for AGV Navigation

A motion planning method that integrates global planning and local control is proposed in this paper, aiming to enable Automated Guided Vehicle (AGV) to navigate automatically in two-dimensional

Vision based navigation in a dynamic environment

A visual servoing controller able to make the robot move over a long distance thanks to a topological map and to the choice of suitable targets and a control law based on equiangular spirals has been designed to guarantee non collision.



A Feedforward Control Approach to the Local Navigation Problem for Autonomous Vehicles

The state space model is described which forms the core technology of an integrated autonomous navigation system incorporating perception, control, and position estimation called RANGER and trajectory tracking, obstacle avoidance and sensor stabilization, aspects of the system are concentrated on.

Rapid physics-based rough-terrain rover planning with sensor and control uncertainty

A model-based planning method is presented in this paper that is computationally efficient and takes into account uncertainty in the robot model, terrain model, range sensor data, and rover path following errors.

Optimal and efficient path planning for partially-known environments

  • A. Stentz
  • Computer Science
    Proceedings of the 1994 IEEE International Conference on Robotics and Automation
  • 1994
A new algorithm, D*, is introduced, capable of planning paths in unknown, partially known, and changing environments in an efficient, optimal, and complete manner.

State Space Sampling of Feasible Motions for High Performance Mobile Robot Navigation in Highly Constrained Environments

This paper presents an effective algorithm for state space sampling based on a model-based trajectory generation approach that enables high-speed navigation in highly constrained and/or partially known environments such as trails, roadways, and dense off-road obstacle fields.

Outdoor Mapping and Navigation Using Stereo Vision

This work considers the problem of autonomous navigation in an unstructured outdoor environment, and uses stereo vision as the main sensor to use more distant objects as landmarks for navigation, and to learn and use color and texture models of the environment.

Rough Terrain Visual Odometry

An integrated system to localize a mobile robot in rough outdoor terrain using visual odometry through the use of bundle adjustment over multiple frames to reduce the error significantly, thereby making the system more robust and accurate while still operating in real-time.

An Interpolated Dynamic Navigation Function

The theoretical basis and a working implementation of the E* algorithm are presented, as well as measurements of the algorithm’s precision, topological correctness, and computational effort.

The dynamic window approach to collision avoidance

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.

Learning to predict slip for ground robots

A generic nonlinear regression framework is proposed in which the terrain type is determined from appearance and then a nonlinear model of slip is learned for a particular terrain type.

Planning algorithms

This coherent and comprehensive book unifies material from several sources, including robotics, control theory, artificial intelligence, and algorithms, into planning under differential constraints that arise when automating the motions of virtually any mechanical system.