ZAPT's Autonomous Engineering & Robotic Solutions
For Repetitive Commercial Tasks
Can My Equipment Be Autonomous?
For ZAPT to develop a solution for your equipment to run autonomously, the machine must perform some form of a “repetitive task,” work in a “defined space,” and work within a clearly defined “Operational Design Domain.”
What do we mean by “repetitive task,” “defined space,” and “Operational Design Domain”?
A “repetitive task” involves performing the same operation, perhaps weekly or daily.
A “defined space” is an area that can be clearly defined or mapped. The defined space may be a relative local space (think - automating the painting of line markings on a football field), or this may be an “absolute” space, where the task is completed with respect to an Earth reference frame (think - driving from Houston to Dallas)
“Operational Design Domain,” or ODD, is where we get into the weeds. The ODD is a collection of information that clearly defines the boundaries for the platform's operations. Just a few of these questions are; what is the maximum slope the platform will work on, what are the limits of weather that the platform will work in (rain, snow, fog, etc.), and will it be required to obey traffic signage?
If the above describes your commercial equipment operations, ZAPT can automate most platforms by integrating a custom-designed autonomous sensor suite into your specific existing equipment.
ZAPT has evaluated platforms and operations that include various commercial lawnmowers, snow removal equipment, street sweepers, grass trimmers, line marking equipment, and small tractor units. The fastest way to learn more about autonomous solutions for your equipment and operations is to reach out to us. We will work with you to complete a feasibility study to evaluate your equipment’s autonomous potential. We can quickly determine the cost and complexity required to automate your operations.
Custom Autonomous Examples
Sensor Integration Experience
ZAPT's complex sensor integration knowledge stems from 17 years of experience performing precise navigation and positioning in the Energy sector from ZAPT's parent company Zupt, precisely positioning subsea remotely operated vehicles in up to 4,000m of water.
For the past two years, we have taken this capability and developed Nomad, ZAPT's fully autonomous commercial mower that is available for lease. There are several predominant sensor types we have expertise in integrating for precise localization and perception. These sensor types include GNSS(GPS), Inertial Measurement Units, LiDAR, Stereo Cameras, Radar, Ultrasonic sensors, and Resolvers/Encoders. Keeping safety at the forefront, we integrate these sensors cost-effectively to offer reliability and redundancy during autonomous operations.
GNSS, more commonly known as GPS, is the most widely used and affordable positioning tool. We use specific configurations of GNSS (RTK, PPP, etc.) to deliver precise localization for autonomous positioning accuracy when data is available, i.e., when we have a clear view from the platform to the orbiting satellites.
An Inertial Measurement Unit (IMU) measures rotational rate and acceleration in three axes. IMUs can be expensive, or they can be affordable. Affordable IMUs correctly integrated (coupled) with other sensors significantly help enable reliable localization and perception solutions.
LiDAR comes in many forms, long-range 360-degree scanning or short-range with a limited field of view. All LiDAR systems generate a "point cloud" or map of the system's space. We use LiDAR for positioning or map matching and object identification as part of the perception solution.
Stereo cameras allow us to identify objects (people, signs, vehicles) and detect the distance to those objects. Using 3D reconstruction techniques with stereo cameras enables us to generate a map of our environment, which we use to aid all other mapping solutions.
Radar has many different sensor types available. The industry has defined the segmentation as Long Range Radar (LRR), Medium (MRR), and Short Range Radar (SRR). These sensors can be more affordable than others, and because they deliver information at a longer distance, they are more appropriate on faster-moving platforms.
Ultrasonic sensors (parking or backup sensors on your car) detect close-range objects by sending out simple ultrasonic pulses reflected by close surrounding objects. These are short-range and affordable sensors.
Resolvers and encoders are used to detect either rotation or extension/retraction. They are critical for accurately determining such things as steering angle or wheel traction. For reliable autonomous operations, all ground drive wheels will need some form of high-resolution encoder/resolver.
We are not limited to the sensors listed above. Depending on the machine platform and application, we will customize the sensor suite to provide a safe, reliable autonomous solution.
Levels of Automation
Level 5 | Full Automation
Performs all tasks under all conditions – zero human attention or interaction required
Level 4 | High Automation
Performs all tasks under specific conditions – geofencing required, human override optional
Level 3 | Conditional Automation
Environmental detection capabilities - must be fully monitored
Level 2 | Partial Automation
Human operator with autonomous steering, line/lane following, and acceleration assistance
Level 1 | User Assistance
Human operator with one level of assistance (automated braking or line/lane assistance)
Level 0 | No Autonomy
Full manual control by user
The SAE defines six levels of driving automation for the automotive sector, from level 0 of no autonomy to level 6 of full automation performing all tasks under all conditions with zero human attention or interaction.
At ZAPT, we can customize your equipment to reach the highest level of autonomy based on your specific platform, operational space, and operating domain. We select the most appropriate sensors, such as GNSS, IMU, LiDAR, stereo cameras, ultrasonics, and resolvers, and integrate them into your equipment. Sensor selection considers safety for the application, the level of reliability of the integrated solution, and the level of redundancy for all operations.
The selection of the correct sensor suite and their optimal fusion creates a robust localization and perception “brain” that allows your platform to reach full autonomy. Some of the main operational advantages of a well-designed solution would include the following:
Higher levels of production due to more precise position data
An autonomous solution that can complete 100% of the task, not just 30% of the work
In mowing applications, no manual headland path is required
Job planning based on satellite/airborne imagery
Automated job planning
The best “bang for the buck” from the sensor suite