ROS Autonomous Control Package
Below, we have laid out different hardware and software options to integrate into a ROS powered robot. Choosing the correct sensors to supply positioning or obstacle detection, given the platform’s environment, will make navigation and control implementation a much smoother process. For a more in-depth guide to selecting the right configuration can be found on our autonomous support page. For an introduction to ROS, please follow these tutorials. If the individual components are the only desired items, this isn't the package for you. This is meant to be a setup and ready-to-go ROS system using the support and services of SuperDroid Robots.
- Required Package Items:
- Motor encoders are a required odometry source for this system. A ROS node will be provided to communicate with a RoboteQ motor controller through USB. A RoboteQ is not included in this package.
- All power distribution and voltage regulator installation is included.
- The UM7 inertial measurement unit (IMU) is required due to its precision, flexibility, and the general necessity of IMU data.
- WiFi is the required communication method with this system. If your application demands the robot to go further than 80 meters from the nearest wireless access point (WAP), then our tactical IP radios will be a good option. For availability and pricing please contact us. Also, please note that an autonomous robot doesn't need to be connected to a network at all times. It will still perform the task. It can't be remotely observed or overridden however.
- Source code includes launch files (XML) with custom ROS nodes (C++). The source code will be provided upon purchase of the package.
- The ROS ReadMe file will include explanations of our custom nodes, reference links to the nodes used from ros.org, and a flow chart showing how the ROS structure is communicating.
- Choose between multiple Linux single board computers:
- The selected computer will come with a full installation of ROS with nodes to support the sensor configuration. This includes thorough testing with the implementation of the chosen control method.
- The Odroid can handle single LIDAR mapping. If an application demands more features, a Jetson TX1/TX2/Xavier is required.
- Choose between our LIDAR options:
- We task the Hokuyo LIDARs with identifying both fixed and moving obstacles in the environment and their location. When paired with a Simultaneous Localization and Mapping (SLAM) algorithm, the robot can simultaneously create a map of an unknown environment using sensor feedback and position the robot within the map on the fly.
- This cost includes ROS setup, testing, and verifying the LIDAR output is suitable for autonomy.
- When selecting different configurations, keep in mind most LIDARs will only provide 2D awareness. Obstacles above or below the LIDAR plane could pose a threat.
- Please note: the only outdoor rated LIDARs are the UTM-30LX-EW and VLP16. If you plan on navigating outdoors, these will be the safest options in order to avoid erroneous data.
- Choose between our two Garmin GPS options:
- If an autonomous robot dares to explore the outdoors, a GPS is something to strongly consider. In vast outdoor areas, GPS modules can provide position feedback where LIDARs can’t. Our GPS options below are more than qualified for the task.
- Both have Wide Area Augmentation System (WAAS) capability which provides up to 3 meter accuracy. Please note that WAAS will only work in the North America region. Expect around 7-10 meter accuracy otherwise.
- When selecting a GPS, you will also be provided with the setup. This includes RS232 to TTL conversion with level shifting to interface with the NVIDIA Jetson's UART communication line. Furthermore, the GPS output will be tested and validated to ensure a WAAS correctional factor is being applied.
- If a more accurate GPS is required, we have experience setting up Real Time Kinematic (RTK) systems to provide 1 to 5 centimeter accuracy. Contact us for more details and pricing
- Check out our sensor support page. for a more in depth comparison of these devices.
- Select a 3D vision option:
- 3D vision systems such as the Xbox Kinect and Zed stereo camera provide obstacle detection where the LIDARs are blind. More often than not, this is a must for robots that have a tall profile or are subject to uneven terrain.
- Purchasing either of these 3D systems includes installation and integration of the necessary software development kit (SDK) and drivers. Additionally, thorough testing will ensure the system provides 3D obstacle detection.
- If this is purchased without a robot platform, you will be responsible for mounting the vision system in an ideal position while adjusting the corresponding field of view parameters (FOV). This is critical for providing ROS with the obstacle detection your robot needs to navigate safely.
- Due to its use of infrared light, the Xbox Kinect will not be suitable for outdoor applications. The Zed will be a better choice that also provides an odometry source.
- ROS Visualization (RViz) is the most popular graphical user interface (GUI) used with ROS. If advanced features such as waypoint queuing and map modifications are desired, select our custom Windows GUI option. This is more geared to end users where RViz is primarily for developers. Special instructions can be entered below.
- Select a level of robot and ROS integration:
- The Manual Control option represents a single board computer with ROS installed that includes all of the necessary code for stable sensor readouts, static coordinate frame transforms, and RoboteQ communication with manual controls. It will be up to you to implement the correct transform offsets and autonomous control methods. The included Logitech wired gamepad will be used for manual control.
- The proceeding checkboxes represent using the sensor data to position the robot in its environment and make it aware of surrounding obstacles. Each sensor presents its own challenges when trying to provide localization. If you plan to select the autonomous navigation option, some of the positioning and detection options need to be checked. Also, these options should not be selected if the corresponding sensor isn't chosen as well.
- The third option is for those who want a complete, ready-to-go autonomous system. This includes control tuning, path planning, and waypoint navigation. Please note if no LIDARs or 3D vision options are selected, the system will not have obstacle avoidance. Otherwise, the path planning algorithm will detect obstacles and route the robot in the most efficient path.
- By default, 1 hour of support is provided for troubleshooting and design consultation. Additional support time is offered in the options below.
- This is a custom product that must be assembled and programmed. Ship date varies with option selection.
- Need expedited shipping? Please contact us today.
- We will contact you to review the details of your order.
This package provides a completely assembled and configured Robot Operating System (ROS). ROS is a Linux based meta operating system for your robot with a massive open source community. It's flexibility allows compatibility with many standalone Linux machines such as the Nvidia Jetson products. For developers or hobbyists, ROS provides low level device control, tools, and libraries for obtaining, building, writing, and running code. Our goal is to provide a ready to go configured system to give your project a head start. As always, please post on our forums or contact us with your questions.
ROS Autonomous Control Package
- Product Code: TE-900-007
- Brand: SuperDroid Robots
- Availability: Built to Order 10-16 weeks
Available Options( Collapse All )
- Logitech USB GamePad Controller
- Encoder Interfacing
- Voltage Regulators and Power Distribution Hardware
- RoboteQ Motor Controller Hookup Hardware
- WiFi Communication
- Source Code
- Setup, Configuration, and Testing
- ROS Documentation
- Manual Control