Autonomous Solutions, Inc. (ASI) has improved an algorithm for autonomous vehicles to detect drop-offs and other large negative obstacles often found in the environments in which automated off-road vehicles operate.

“ASI has developed a method for mapping point cloud occlusions in real-time,” said Taylor Bybee, Perception Tech Lead at ASI. “Which provides additional accuracy and safety when integrated into an autonomous vehicle obstacle detection and avoidance system.”

For safe navigation through an environment, autonomous ground vehicles rely on sensor data representing 3D space surrounding the vehicle. Often this data is obscured by objects or terrain, producing gaps in the sensor field of view. These gaps, or occlusions, can indicate the presence of obstacles, negative obstacles, or rough terrain.

Occlusions can be defined as a blockage which prevents a sensor from gathering data in a location. For example, occlusions can be seen as shadows in LiDAR data.

Because sensors receive no data in these occlusions, sensor data provides no explicit information about what might be found in the occluded areas. Information about the occlusions must be inferred from using an occlusion mapping algorithm to provide the navigation system with a more complete model of the environment.

“While sensor data itself doesn’t tell us what’s in the occluded areas, occlusions can represent negative obstacles like drop-offs or areas behind large obstacles,” said Jeff Ferrin, CTO at ASI. “It’s important to identify these areas for obstacle detection and avoidance to work properly.”

Application of this new technology can be useful in settings with dump edges at mine sites, steep road edges, canals, ditches, hills or stairs for indoor or urban environments.

The occlusion mapping algorithm has three main components.

The first is a sensor field of view (FOV) model that describes what obstacles the sensors are expected to detect. This component is designed for point cloud sensors such as 3D LiDAR, Flash LiDAR, Structured Light, and Stereo Cameras.

Second, an occlusion map is maintained and updated using the sensor FOV model and current sensor data to provide a probabilistic estimate on areas that have not been detected within the sensor FOV.

The third component is the integration of the occlusion map into an autonomous vehicle navigation system. It is designed to work with and complement existing obstacle detection and avoidance systems.

About ASI

Autonomous Solutions, Inc. (ASI) is a world leader in industrial vehicle automation. ASI serves clients across the world in the mining, agriculture, automotive, government, and manufacturing industries with remote control, teleoperation, and fully automated solutions from its headquarters and 100-acre proving ground in northern Utah.

Autonomous Solutions, Inc. (ASI) and CNH Industrial have announced the unveiling of concept autonomous tractors. ASI is CNH Industrial’s technology provider responsible for developing autonomous vehicle technology for a concept cabless Case IH Magnum and a concept New Holland T8, based on a current production tractor.

The tractors, unveiled today at the Farm Progress Show in Boone, Iowa, include the ability for autonomous seeding, planting and tillage, for broad acre and row crop farming. Advanced path planning technology will allow farmers to manage their fields efficiently and oversee the overall operation of several vehicles at once. The vehicles are also capable of obstacle detection which will enhance safety in the agriculture industry.

“ASI is the industry-leading developer of autonomous solutions in the off-road vehicle segment"

“ASI is the industry-leading developer of autonomous solutions in the off-road vehicle segment and the natural choice to be one of CNH Industrial’s key technology providers. CNH Industrial has had a long and successful relationship with ASI and we will continue to work together in developing advanced autonomous solutions for the future benefit of our global customers” stated Richard Tobin, CEO of CNH Industrial.

This new technology gives farmers the ability to operate their tractor as both a manned or unmanned vehicle.

At a farmer's command, autonomous tractors can drive from a parking area along private roads to a field and begin work without any intervention.

ASI has nearly two decades of autonomous technology development experience. As a smaller and more agile technology developer, ASI is able to partner with large global companies to help them disrupt their market with multi-vehicle autonomy faster and more economically than they could in any other way.

“Our relationship with CNH Industrial is vital in facilitating the near term disruption of how farming is done. We’re thrilled to be working with the leaders in Ag innovation to make this exciting future of driverless tractors a reality,” says Mel Torrie, ASI founder and CEO. “CNH Industrial’s selection of ASI as a long term, strategic robotic development provider validates the capability and flexibility of our robotics platform in reducing the risk and costs for equipment manufacturers to bring advanced capabilities to their respective industries.”

“Our relationship with CNH Industrial is vital in facilitating the near term disruption of how farming is done."

ASI and CNH Industrial have joined forces to create a development model and architecture framework that is flexible and dynamic, able to quickly adapt and adopt new technologies and standards as fast as they become available. This concept tractor results from the integration of ASI’s autonomous hardware and software with CNH Industrial’s advanced platform.

ASI also leverages this autonomous technology with other large global companies such as FCA US, Ford, Toyota, Rio Tinto, Anglo American, Boeing, Lockheed Martin, Northrop Grumman, Dematic, and a large global security company and others to be announced in coming weeks.

Want More Info?

Watch the Video Here

About Autonomous Solutions, Inc.

Autonomous Solutions, Inc. (ASI) designs and manufactures unmanned vehicle systems, software, and components that are OEM agnostic. ASI’s robotic kits and command and control software can be leveraged in a variety of industries including agriculture, automotive, mining, military, security, and industrial applications. ASI’s vehicle automation kits are OEM independent and have been used for many different applications. Clients have experienced enhanced safety, increased efficiency, and improved accuracy.

For over 16 years, ASI has been a leader in vehicle automation. ASI has a strong software team which makes up over half of its engineering work force. ASI’s Mobius software allows for command and control of a fleet of vehicles in addition to ease of use, industry specific task planning, and vehicle diagnostics. ASI has extensive experience automating vehicles of all shapes and sizes with an emphasis on systems integration, obstacle detection and avoidance, tele-operation, and fully autonomous controls.

About CNH Industrial N.V.

CNH Industrial N.V. (NYSE: CNHI /MI: CNHI) is a global leader in the capital goods sector with established industrial experience, a wide range of products and a worldwide presence. Each of the individual brands belonging to the Company is a major international force in its specific industrial sector: Case IH, New Holland Agriculture and Steyr for tractors and agricultural machinery; Case and New Holland Construction for earth moving equipment; Iveco for commercial vehicles; Iveco Bus and Heuliez Bus for buses and coaches; Iveco Astra for quarry and construction vehicles; Magirus for firefighting vehicles; Iveco Defence Vehicles for defence and civil protection; and FPT Industrial for engines and transmissions. More information can be found on the corporate website: www.cnhindustrial.com

Media Contact

Matt Nielsen
Corporate Communications Manager

At the end of last year, ten of the world's largest consumer vehicle manufacturers combined to announce that automated emergency braking systems would be a standard feature in all new vehicle models produced by the companies. The move is designed to reduce the number of rear end collisions which make up about a third of all accidents.

and autonomous braking system braking for a vehicle

While each manufacturer's system would work differently, the fundamental concept is the same: if forward facing sensors detect a slower moving or stopped vehicle ahead, the brakes automatically engage without driver intervention. "We are entering an era of vehicle safety, focused on preventing crashes from ever occurring, rather than just protecting occupants when crashes happen," said Anthony Foxx, US Transportation Secretary in the Los Angeles Times article.

"We are entering an era of vehicle safety, focused on preventing crashes from ever occurring, rather than just protecting occupants when crashes happen,"

Regardless of where your opinion falls on this specific topic, it's important to recognize that the automatic emergency braking system is just one of several autonomous vehicle technologies that have or will shortly make it into consumer vehicles. Advances such as adaptive cruise control, lane departure warnings, and automatic parallel parking have already made their debut and are becoming increasingly affordable options. In each case, the combination of sensors and intelligent action on the part of the vehicle has its roots in robotics. The benefits to safety and convenience will continue to push the industry toward robotic vehicles.

However, because of the substantial legal issues involved, the unmanned industry generally has steered clear of consumer vehicles. California recently released draft regulation proposals signaling that liability of these vehicles could rest on the vehicle manufacturers. Nothing has been passed on the federal or state level though so some are still trying to anticipate the eventual adoption of fully automated vehicles on public roads and what it will take to get the industry there. For example, the University of Michigan, in partnership with the Michigan DOT, have created Mcity, a mock "downtown" that simulates the unexpected variables that come with driving in urban conditions, allowing manufacturers to test autonomous vehicles in a realistic environment. ASI also expanded its test facilities in northern Utah to allow for simulating some of these situations.

Fully autonomous systems have already been successfully applied to consumer vehicles in closed/controlled environments. Autonomous Solutions, Inc. (ASI) has been working with Ford Motor Company since 2011 and developed a robotic durability testing program. Robotic technology is retrofitted to new vehicle models and operated from a central command center at Ford's Michigan Proving Ground facility. The robotic technology enables Ford to remove human drivers from the most punishing test tracks and run durability testing 24/7.

Despite the progress of the University of Michigan, ASI, and other innovators like Google with their Google Car program, full adoption of autonomous vehicles on the road still seems a ways in the future. But that's not to say that we won't continue to see individual autonomous technologies, such as automatic emergency braking, trickle into the consumer market.

Raw point cloud data generated by ASI's Forcast lidar and analyzed by Vantage™ perception softwareRaw point cloud data generated by ASI's Forcast lidar and analyzed by Vantage™ perception software

Each autonomous vehicle system employs a set of sensors that provide environmental feedback for safety, navigation, and other essential vehicle functions. Up until this point, sensors with high levels of accuracy have been too expensive for most users to afford, but that is rapidly changing.

From a vertically spinning LIDAR on a mining truck to a stereo camera on a Packbot , it's not uncommon to see any one of a number of sensors atop an unmanned vehicle at Autonomous Solutions, Inc. (ASI). GPS, LIDAR, infrared, and vision-based sensor technologies play an integral role in any autonomous vehicle system. Feedback from a network of sensors and software algorithms enables an unmanned vehicle to essentially “see" what is going on in its environment and react intelligently.

Unfortunately, sensors--which are generally the largest cost-driver in the autonomous system--and budgets do not always match up, forcing companies to sacrifice their desired level of accuracy for a more affordable sensor. Change is close, however, according to ASI's Jared Pratt.
"As sensor prices have dropped and as the algorithms have improved to leverage lower cost sensors, more automation applications are showing a strong return on investment"
said Pratt in a recent article by OEM-Off Highway's Sara Jensen.

Safety is ASI's number one priority. Requesting a safety audit from HORIBA MIRA helps ASI evaluate and enhance its internal processes to ensure that rigorous international safety standards are continually in place to be met.Safety is ASI's number one priority. Requesting a safety audit from HORIBA MIRA helps ASI evaluate and enhance its internal processes to ensure that rigorous international safety standards are continually in place to be met.

ASI is pleased to have been audited by an external auditor Dr. David Ward of HORIBA MIRA. ASI engineering processes were reviewed over a period of 4 days this summer and the approach to the functional safety of our products was assessed against a range of international functional safety engineering standards including IEC 61508, ISO 26262, ISO 17757 and ISO 13849.

“It’s a pleasure working with HORIBA MIRA and David Ward. He’s a professional of the utmost integrity and patience, always pragmatic and extremely knowledgeable in a broad range of sectors including Mining, Security, Agriculture and Automotive where ASI customers are realizing huge savings and productivity improvements through using our products and services,” says Jonathan Moore, ASI Chief Engineer.

“As autonomous solutions become more widespread in automotive and industrial applications it’s important that we demonstrate their dependability, compliance with functional safety standards is an important aspect of demonstrating a rigorous approach to product design and implementation,” says Dr. David Ward, HORIBA MIRA.

“It’s encouraging to see a technology innovator embodying these principles as part of their core engineering processes.”

8412. The number of work-related injuries reported in 2012 in the mining industry. While that number is down from previous years, clearly there is still room to improve safety in one of the most dangerous industries in the world.

"There's a lot of focus on mining safety," says Mel Torrie, CEO of Autonomous Solutions, Inc. "Mine workers are under constant threat of injury from falling objects, unstable terrain, and working with heavy machinery." Torrie, who will speak about safety at the RobotXWorld conference in late August, identifies safety as ASI's number one concern. "Vehicle automation is first and foremost a safety technology."

Each autonomy kit developed by ASI delivers four layers of safety.

At the basic level, all automated vehicles are equipped with emergency stop functionality. If a vehicle travels outside of a designated operation area, experiences a malfunction, or loses communications, the emergency stop system is initiated to halt vehicle operations until the issue can be resolved.

The on-board computer, called the vehicle control unit (VCU), serves as the brain of the automated system, relaying commands from the remote operator to the vehicle functions, monitors vehicle health indicators, and relaying data packets back to the operator.

The VCU represents the next level of safety as it actively runs more than 100 system error checks looking for errors that might cause safety issues.

While autonomous vehicles may accurately navigate by following GPS waypoints, they are still "blind" in the sense that they are unable to see potentially hazardous obstacles in their path such as animals, people, and other vehicles.

To solve this safety issue, ASI equips autonomous vehicles with obstacle detection and avoidance software and sensors that actively search for obstacles within the path of the vehicle and warn the VCU and operator of their presence. Depending on the installation, vehicles may react to obstacles with see-and-stop or see-and-avoid behavior.

The final layer of safety is ASI's command and control software, Mobius™. Operators use the Mobius path builder to determine a safety perimeter for safe autonomous operations.

If for any reason a vehicle travels outside of this perimeter, the emergency stop system halts autonomous operations until the vehicle is moved (manually or by teleoperation) back inside the safety perimeter. For systems that manage more than one vehicle, Mobius actively monitors the health indicators of all vehicles in the system and initiates proximity alerts that slow down vehicles if they are in danger of colliding.

35. The number of mining related deaths reported in 2012. That's 35 too many. With four layers of redundant safety features, ASI is setting the safety standard for autonomous vehicles that will help drive down injuries and fatalities across all industries. "Our goal is to supply our customers with solutions that not only help them improve productivity but also dramatically improve the safety of their work environment," said Torrie.