Category Archives: Automation & autonomous systems

U.S. Development of Lethal Autonomous Weapon Systems

Peter Lobner

Introduction

In November 2022, the Congressional Research Service (CRS) published an update to their document, “Defense Primer: U.S. Policy on Lethal Autonomous Weapon Systems,” which is available on the CRS website here: https://s3.documentcloud.org/documents/23310494/if11150.pdf

Each of the US military services has its own autonomous vehicle / weapons system programs.  Following is a brief roadmap to those programs.

Navy

In November 2022, CRS published an update to their document, “Navy Large Unmanned Surface and Undersea Vehicles: Background and Issues for Congress,” which you can read here: https://s3.documentcloud.org/documents/21271730/navy-large-unmanned-surface-and-undersea-vehicles-background-and-issues-for-congress-feb-17-2022.pdf

See my April 2016 post, “Large Autonomous Vessels will Revolutionize the U.S. Navy,” for background information on the Navy’s autonomous vessel program and the Sea Hunter prototype developed by Leidos and tested in San Diego: https://lynceans.org/tag/continuous-trail/

The Navy’s San Diego-based Unmanned Surface Vessel Division One is playing an important role in developing and testing several autonomous vessels.

Medium displacement unmanned surface vessels Seahawk (front) and Sea Hunter leave San Diego Bay ahead of the large manned destroyer USS Zumwalt.
Source: USNI 2021

For more information on the Navy’s autonomous vessel program, check out these US Naval Institute articles:

Air Force

In July 2022, CRS provided an overview of unmanned and autonomous aerial system in their report, “Unmanned Aircraft Systems: Roles, Missions, and Future Concepts,” which you’ll find here: https://crsreports.congress.gov/product/pdf/R/R47188

In one program, the USAF now is testing an AI-controlled F-16 fighter aircraft in aerial combat scenarios.  Here’s the February 2023 story: https://www.military.com/daily-news/2023/02/22/fully-autonomous-f-16-fighter-jet-takes-part-dogfights-during-tests.html?ESRC=eb_230223.nl

In another program, the USAF is developing an autonomous “wingman” aircraft to fly along with manned fighter aircraft to provide greater capabilities to attack a target and/or provide protection for the manned aircraft.  This development is described in a February 2022 article on The Drive website here: https://www.defensenews.com/air/2022/02/13/how-autonomous-wingmen-will-help-fighter-pilots-in-the-next-war/

Defense contractor Kratos, which has offices in San Diego, has important roles in several DoD autonomous aerial systems projects.

An XQ-58A Valkyrie unmanned aerial vehicle flies in formation with an F-22 Raptor and F-35A Lightning over the U.S. Army Yuma Proving Ground testing range, Ariz., during a series of tests in Dec. 2020. Source: USAF photo

Army

In January 2023, CRS published an update to their document, ”The Army’s Robotic Combat Vehicle (RCV) Program,” which is available here: https://crsreports.congress.gov/product/pdf/IF/IF11876

Army’s Robotic Combat Vehicle RCV-M (medium) prototype. Source: CRS

As you can see, there’s a lot going on in this field and capabilities for use of lethal autonomous systems may soon challenge limits set by present policy.

What Do You Put On The Borg Warner Trophy When An Autonomous Car Wins the Indy 500?

Peter Lobner

A year ago, this might have seemed like a foolish question.  An autonomous car racing in the Indianapolis 500 Mile Race?  Ha!  When pigs fly!

The Indy 500 Borg Warner Trophy. 
Source:  The359 – Flickr via Wikipedia

One of the first things you may notice about the Borg Warner Trophy is that the winning driver of each Indy 500 Race is commemorated with a small portrait/sculpture of their face in bas-relief along with a small plaque with their name, winning year and winning average speed. Today, 105 faces grace the trophy.

Borg Warner Trophy close-up.
Source: WISH-TV, Indianapolis, March 2016

The Indianapolis Motor Speedway (IMS) website provides the following details:

“The last driver to have his likeness placed on the original trophy was Bobby Rahal in 1986, as all the squares had been filled. A new base was added in 1987, and it was filled to capacity following Gil de Ferran’s victory in 2003. For 2004, Borg-Warner commissioned a new base that will not be filled to capacity until 2034.”

On 11 January 2021, the Indianapolis Motor Speedway along with Energy Systems network announced the Indy Autonomous Challenge (IAC), with the inaugural race taking place at the IMS on 23 October of 2021.  The goal of the IAC is to create the fastest autonomous race car that can complete a head-to-head 50 mile (80.5 km) race at IMS. The challenge, which offers $1.5 million in prize money, is geared towards college and university teams. The IAC website is here: https://www.indyautonomouschallenge.com

The IAC organizers state that this challenge was “inspired and advised by innovators who competed in the Defense Advanced Research Projects Agency (DARPA) Grand Challenge, which put forth a $1 million award in 2004 that created the modern automated vehicle industry.”

All teams will be racing an open-wheel, automated Dallara IL-15 race car that appears, at first glance, quite similar to conventional (piloted) 210 mph Dallara race cars used in the Indy Lights race series.  However, the IL-15 has been modified with hardware and controls to enable automation.  The automation systems include an advanced set of sensors (radar, lidar, optical cameras) and computers.  Each completed race car has a value of more than $1 million. The teams will focus primarily on writing the software that will process the sensor data and drive the cars.  When fully configured for the race, the IAC Dallara IL-15 will be the world’s fastest autonomous automotive vehicle.

Rendering of the autonomous Dallara IL-15.  Source: IAC
Rendering of the autonomous Dallara IL-15 on the IMS race track.  Source: IAC

Originally, 39 university teams from 11 counties and 14 states had applied to compete in the IAC.  As of mid-January 2021, the IAC website lists 24 teams still actively seeking to qualify for the race.  

The race winner will be the first team whose car crosses the finish line after a 20-lap (50 mile / 80.5 km) head-to-head race that is completed in less than 25 minutes.  This requires an average lap speed of at least 120 mph (193 kph) and an average lap time of less than 75 seconds around the 2.5 mile (4 km) IMS race track. 

In comparison, Indy Light races at IMS from 2003 to 2019 have had an average winning speed of 148.1 mph (238.3 kph) and an average winning lap time of 60.8 seconds.  All of these races were run with cars using a Dallara chassis. The highest winning average speed for an Indy Lights race at IMS was in 2018, when Colton Herta won in a Dallara-Mazda at an average speed of 195.0 mph (313.8 kph) and an average lap time of 46.1 seconds, with no cautions during the race.

Milestones preceding the autonomous race are listed on the IAC website here: https://www.indyautonomouschallenge.com/timeline

Key milestones include:

  • 27 – 29 May: Vehicle distribution to the teams
  • 5 – 6 June: Track practice #1
  • 4 – 6 September: Track practice #2
  • 19 – 20 October: Track practice #3
  • 21 – 22 October: Final race qualification
  • 23 October: Race day

The winning team will receive a prize of $1 million, with the second and third place teams receiving $250,000 and $50,000, respectively.

The IAC race will be held more than 17 years after the first of three DARPA Grand Challenge autonomous vehicle competitions that were instrumental in building the technical foundation and developing broad-based technical competencies related to autonomous vehicles.  A quick look at these DARPA Grand Challenge races may help put the upcoming IAC race in perspective.

The first DARPA Grand Challenge autonomous vehicle race was held on 13 March 2004.  From an initial field of 106 applicants, DARPA selected 25 finalists. After a series of pre-race trials, 15 teams qualified their vehicles for the race. The “race course” was a 140 mile (225 km) off-road route designated by GPS waypoints through the Mojave Desert, from Barstow, CA to Primm, NV.  You might remember that no vehicles completed the course and there was no winner of the $1 million prize. The vehicle that went furthest was the Carnegie Mellon Sandstorm, a modified Humvee sponsored by SAIC, Boeing and others.  Sandstorm broke down after completing 7.36 miles (11.84 km), just 5% of the course. 

A second Grand Challenge race was held 18 months later, on 8 October 2005. DARPA raised the prize money to $2 million for this 132 mile (212 km) off-road race. From an original field of 197 applicants, 23 teams qualified to have their vehicles on the starting line for the race.  In the end, five teams finished the course, four of them in under the 10-hour limit. Stanford University’s Stanley was the overall winner.  All but one of the 23 finalist teams traveled farther than the best vehicle in 2004.  This was a pretty remarkable improvement in autonomous vehicle performance in just 18 months.

In 2007, DARPA sponsored a different type of autonomous vehicle competition, the Urban Challenge.  DARPA describes this competition as follows:

“This event required teams to build an autonomous vehicle capable of driving in traffic, performing complex maneuvers such as merging, passing, parking, and negotiating intersections. As the day wore on, it became apparent to all that this race was going to have finishers. At 1:43 pm, “Boss”, the entry of the Carnegie Mellon Team, Tartan Racing, crossed the finish line first with a run time of just over four hours. Nineteen minutes later, Stanford University’s entry, “Junior,” crossed the finish line. It was a scene that would be repeated four more times as six robotic vehicles eventually crossed the finish line, an astounding feat for the teams and proving to the world that autonomous urban driving could become a reality. This event was groundbreaking as the first time autonomous vehicles have interacted with both manned and unmanned vehicle traffic in an urban environment.”

In January 2021, a production Tesla Model 3 with the new Full Self-Driving (FSD) Beta software package drove from San Francisco to Los Angeles with almost no human intervention.  I wonder how that Tesla Model 3 would have performed on the 2007 DARPA Urban Challenge.  You can read more about the SF – LA FSD trip at the following link: https://interestingengineering.com/tesla-full-self-driving-successfully-takes-model-3-from-sf-to-la

We’ve seen remarkable advances in the development of autonomous vehicles in the 17 years since the 2004 DARPA Grand Challenge race.  Is it unreasonable to think that an autonomous race car will become competitive with a piloted Indy race car during the next decade and compete in the Indy 500 before they run out of space on the Borg Warner Trophy in 2034?  If the autonomous racer wins the Indy 500, what will they put on the trophy to commemorate the victory? A silver bas-relief of a microchip?

I think I see a flying pig!

For more information on IAC and IMS

For more information on the DARPA Grand Challenges for autonomous vehicles

Festo’s SmartBird and BionicSwift – A Decade of Progress in Deciphering How Birds Fly

Peter Lobner

1. Background on Festo

Festo is a German multinational industrial control and automation company based in Esslingen am Neckar, near Stuttgart. The Festo website is here: https://www.festo.com/group/en/cms/10054.htm

Festo reports that they invest about 8% of their revenues in research and development.  Festo’s draws inspiration for some of its control and automation technology products from the natural world. To help facilitate this, Festo established the Bionic Learning Network, which is a research network linking Festo to universities, institutes, development companies and private inventors.  A key goal of this network is to learn from nature and develop “new insights for technology and industrial applications”…. “in various fields, from safe automation and intelligent mechatronic solutions up to new drive and handling technologies, energy efficiency and lightweight construction.”

One of the challenges taken on by the Bionic Learning Network was to decipher how birds fly and then develop robotic devices that can implement that knowledge and fly like a bird. Their first product was the 2011 SmartBird and their newest product is the 2020 BionicSwift.  In this article we’ll take a look at these two bionic birds and the significant advancements that Festo has made in just nine years.

2. SmartBird

On 24 March 2011, Festo issued a press release introducing their SmartBird flying bionic robot, which was one of their 2011 Bionic Learning Network projects. Festo reported:

  • “The research team from the family enterprise Festo has now, in 2011, succeeded in unraveling the mystery of bird flight. The key to its understanding is a unique movement that distinguishes SmartBird from all previous mechanical flapping wing constructions and allows the ultra-lightweight, powerful flight model to take off, fly and land autonomously.”
  • “SmartBird flies, glides and sails through the air just like its natural model – the Herring Gull – with no additional drive mechanism. Its wings not only beat up and down, but also twist at specific angles. This is made possible by an active articulated torsional drive unit, which in combination with a complex control system makes for unprecedented efficiency in flight operation. Festo has thus succeeded for the first time in attaining an energy-efficient technical adaptation of this model from nature.”

SmartBird measures 1.07 meters (42 in) long with a wingspan of 2.0 meters (79 in) and a weigh of 450 grams (16 ounces, 1 pound).  This is about a 1.6X scale-up in the length and span of an actual Herring Gull, but at about one-third the weight. It is capable of autonomous takeoff, flight, and landing using just its wings, and it controls itself the same way birds do, by twisting its body, wings, and tail.  SmartBird’s propulsion system has a power requirement of 23 watts.

Source:  All three SmartBird photos from Festo

More information on SmartBird is on the Festo website here:  https://www.festo.com/group/en/cms/10238.htm

You can watch a 2011 Festo video, “Festo – SmartBird,” (1:47 minutes) on YouTube here:  https://www.youtube.com/watch?v=nnR8fDW3Ilo

3. Bionic Swift

On 1 July 2020, Festo introduced the BionicSwift as their latest ultra light flying bionic robot that mimics how actual birds fly. 

The BionicSwift, inspired by a Common Swift, measures 44.5 cm (17.5 in) long with a wingspan of 68 cm (26.7 in) and a weight of just 42 grams (1.5 ounces). It’s approximately a 2X scale-up of a Common Swift, but still a remarkably compact, yet complex flying machine with aerodynamic plumage that closely replicates the flight feathers on an actual Swift.  The 2011 SmartBird was more than twice the physical size and ten times heavier.

The BionicSwift is agile, nimble and can even fly loops and tight turns.  Festo reports: “Due to this close-to-nature replica of the wings, the BionicSwifts have a better flight profile than previous wing-beating drives.”  Compare the complex, feathered wing structure in the following Festo photos of the BionicSwift with the previous photos showing the simpler, solid wing structure of the 2011 SmartBird.

Source:  All three BionicSwift photos from Festo

A BionicSwift can fly singly or in coordinated flight with a group of other BionicSwifts.  Festo describes how this works: “Radio-based indoor GPS with ultra wideband technology (UWB) enables the coordinated and safe flying of the BionicSwifts. For this purpose, several radio modules are installed in one room. These anchors then locate each other and define the controlled airspace. Each robotic bird is also equipped with a radio marker. This sends signals to the anchors, which can then locate the exact position of the bird and send the collected data to a central master computer, which acts as a navigation system.”  Flying time is about seven minutes per battery charge.

More information on the Bionic Swift is on the Festo website here:  https://www.festo.com/group/en/cms/13787.htm

You also can watch a 2020 Festo video, “Festo – BionicSwift,” (1:45 minutes) on YouTube here: https://www.youtube.com/watch?v=v8fgc77dwwg

4. For more information about other Festo bionic creations: 

I encourage you to visit the Festo BionIc Learning Network webpage at the following link and browse the resources available for the many intriguing projects. https://www.festo.com/group/en/cms/10156.htm

On this webpage you’ll find a series of links listed under the heading  “More Projects,” which will introduce you to the wide range of Bionic Learning Network projects since 2006.

You also can watch the following YouTube short videos of Festo’s many bionic creations:

The Importance of Baseload Generation and Real-Time Control to Grid Stability and Reliability

Peter Lobner

On 23 August 2017, the Department of Energy (DOE) issued a report entitled, “Staff Report to the Secretary on Energy Markets and Reliability.” In his cover letter, Energy Secretary Rick Perry notes:

“It is apparent that in today’s competitive markets certain regulations and subsidies are having a large impact on the functioning of markets, and thereby challenging our power generation mix. It is important for policy makers to consider their intended and unintended effects.”

Among the consequences of the national push to implement new generation capacity from variable renewable energy (VRE) resources (i.e., wind & solar) are: (1) increasing grid perturbations due to the variability of the output from VRE generators, and (2) early retirement of many baseload generating plants because of several factors, including the desire of many states to meet their energy demand with a generating portfolio containing a greater percentage of VRE generators. Grid perturbations can challenge the reliability of the U.S. bulk power systems that comprise our national electrical grid. The reduction of baseload capacity reduces the resilience of the bulk power system and its ability dampen these perturbations.

The DOE staff report contains the following typical daily load curve. Baseload plants include nuclear and coal that operate at high capacity factor and generally do not maneuver in response to a change in demand. The intermediate load is supplied by a mix of generators, including VRE generators, which typically operate at relatively low capacity factors. The peak load generators typically are natural gas power plants that can maneuver or be cycled (i.e., on / off) as needed to meet short-term load demand. The operating reserve is delivered by a combination of power plants that can be reliably dispatched if needed.

The trends in new generation additions and old generation retirements is summarized in the following graphic from the DOE staff report.

Here you can see that recent additions (since 2006) have focused on VRE generators (wind and solar) plus some new natural gas generators. In that same period, retirements have focused on oil, coal and nuclear generators, which likely were baseload generators.

The DOE staff report noted that continued closure of baseload plants puts areas of the country at greater risk of power outages. It offered a list of policy recommendations to reverse the trend, including providing power pricing advantages for baseload plants to continue operating, and speeding up and reducing costs for permitting for baseload power and transmission projects.

Regarding energy storage, the DOE staff report states the following in Section 4.1.3:

“Energy storage will be critical in the future if higher levels of VRE are deployed on the grid and require additional balancing of energy supply and demand in real time.”

“DOE has been investing in energy storage technology development for two decades, and major private investment is now active in commercializing and the beginnings of early deployment of grid-level storage, including within microgrids.”

Options for energy storage are identified in the DOE staff report.

You can download the DOE staff report to the Secretary and Secretary Perry’s cover letter here:

https://energy.gov/downloads/download-staff-report-secretary-electricity-markets-and-reliability

Lyncean members should recall our 2 August 2017 meeting and the presentation by Patrick Lee entitled, “A fast, flexible & coordinated control technology for the electric grid of the future.” This presentation described work by Sempra Energy and its subsidiary company PXiSE Energy Solutions to address the challenges to grid stability caused by VRE generators.   An effective solution has been demonstrated by adding energy storage and managing the combined output of the VER generators and the energy storage devices in real-time to match supply and demand and help stabilize the grid. This integrated solution, with energy storage plus real-time automated controls, appears to be broadly applicable to VRE generators and offers the promise, especially in Hawaii and California, for resilient and reliable electrical grids even with a high percentage of VRE generators in the state’s generation portfolio.

You can download Patrick Lee’s 2 August 2017 presentation to the Lyncean Group of San Diego at the following link:

https://lynceans.org/talk-113-8217/

Wave Glider Autonomous Vehicle Harvests Wave and Solar Power to Deliver Unique Operational Capabilities at Sea

Peter Lobner

The U.S. firm Liquid Robotics, Inc., in Sunnyvale, CA, designs, manufactures, and sells small unmanned surface vehicles (USVs) called Wave Gliders, which consist of two parts: an underwater “glider” that provides propulsion and a surface payload vehicle that houses electronics and a solar-electric power system. The physical arrangement of a Wave Glider is shown in the following diagrams. The payload vehicle is about 10 feet (305 cm) long. The glider is about 7 feet (213 cm) long and is suspended about 26 feet (800 cm) below the payload vehicle.

Wave Glider configurationSource: Liquid Robotics. Note: 800 cm suspension distance is not to scale.

The payload vehicle is topped with solar panels and one or more instrumentation / communication / navigation masts. The interior modular arrangement of a Wave Glider is shown in the following diagram. Wave Glider is intended to be an open, extensible platform that can be readily configured for a wide range of missions.

Wave Glider configuration 2Source: Liquid Robotics

The Wave Glider is propelled by wave power using the operational principle for wave power harvesting shown in the following diagram. Propulsion power is generated regardless of the heading of the Wave Glider relative to the direction of the waves, enabling sustained vehicle speeds of 1 to 3 knots.

Wave Glider propulsion schemeSource: Liquid Robotics

The newer SV3 Wave Glider has a more capable electric power system than its predecessor, the SV2, enabling the SV3 glider to be equipped with an electric motor-driven propeller for supplementary solar-electric propulsion. SV3 also is capable of towing and supplying power to submerged instrument packages.

Autonomous navigation and real-time communications capabilities enable Wave Gliders to be managed individually or in fleets. The autonomous navigation capability includes programmable course navigation, including precise hold-station capabilities, and surface vessel detection and avoidance.

Originally designed to monitor whales, the Wave Glider has matured into a flexible, multi-mission platform for ocean environmental monitoring, maritime domain awareness / surveillance, oil and gas exploration / operations, and defense.

More information and short videos on the operation of the Wave Glider are available on the Liquid Robotics website at the following link:

http://www.liquid-robotics.com/platform/overview/

On 28 April 2016, the U.S. Navy announced that it was in the process of awarding Liquid Robotics a sole-source contract for Wave Glider USV hardware and related services. You can read the Notice of Intent at the following link:

https://www.fbo.gov/index?s=opportunity&mode=form&id=6abb899b3e3286bfcd861fc5dedfdb65&tab=core&_cview=0

As described by the Navy:

“The required USV is a hybrid sea-surface USV comprised of a submerged ‘glider’ that is attached via a tether to a surface float. The vehicle is propelled by the conversion of ocean wave energy into forward thrust, independent of wave direction. No electrical power is generated by the propulsion mechanism.”

Navy requirements for the Wave Glider USV include the following:

  • Mission: Capable of unsupported autonomous missions of up to ten months duration, with long distance transits of up to 1,000 nautical miles in the open ocean
  • Propulsion: Wave power harvesting at all vehicle-to-wave headings, with sustained thrust adequate under own propulsion sufficient to tow significant loads
  • Electric Power: Solar energy harvesting during daylight hours, with power generation / storage capabilities sufficient to deliver ten watts to instrumentation 24/7
  • Instrumentation: Payload of 20 pounds (9.1 kg)
  • Navigation: Commandable vehicle heading and autonomous on-board navigation to a given and reprogrammable latitude/longitude waypoint on the ocean’s surface
  • Survivability: Sea states up to a rating of five and winds to 50 knots
  • Stealth: Minimal radar return, low likelihood of visual detectability, minimal radiated acoustic noise

In my 11 April 2016 post, I discussed how large autonomous surface and underwater vehicles will revolutionize the ways in which the U.S. Navy conducts certain operational missions. Wave Glider is at the opposite end of the autonomous vehicle size range, but retains the capability to conduct long-duration, long-distance missions. It will be interesting to see how the Navy employs this novel autonomous vehicle technology.

Landing a Reusable Booster Rocket on a Dime

Updated 18 March 2020

Peter Lobner

There are two U.S. firms that have succeeded in launching and recovering a booster rocket that was designed to be reusable. These firms are Jeff Bezos’ Blue Origin and Elon Musk’s SpaceX.   Their booster rockets are designed for very different missions.

  • Blue Origin’s New Shepard booster and capsule are intended for brief, suborbital flights for space tourism and scientific research. The booster and capsule will be “man-rated” for passenger-carrying suborbital missions.
  • In contrast, SpaceX’s Falcon 9 booster rocket is designed to deliver a variety of payloads to Earth orbit. The payload may be the SpaceX Dragon capsule or a different civilian or military spacecraft. Currently, the Falcon 9 booster and Dragon capsule are not “man-rated” for orbital missions. SpaceX is developing a crewed version of the Dragon capsule that, in the future, will be used to deliver and return crewmembers for the International Space Station (ISS).

Both firms cite a cost advantage of recovering and reusing an expensive booster rocket and space capsule. Let’s see how they’re doing.

Blue Origin

The basic flight profiles of a single-stage, single engine New Shepard booster and capsule are shown in the following diagram. The primary goals of each flight are to boost the capsule and passengers above 62.1 miles (100 km), safely recover the capsule and passengers, and safely recover the booster rocket. You can see in the diagram that the booster rocket and the capsule separate after the booster’s rocket engine is shutdown and they are recovered separately. At separation, the booster and capsule are traveling at about Mach 3 (about 1,980 mph, 3,186 kph). The orientation of the booster rocket is controlled during descent and the rocket engine is restarted once at low altitude to bring the booster to a soft, vertical landing. Both the booster rocket and the capsule are designed for reuse.

Blue-origin-flight-profileSource: Blue Origin

On 23 November 2015, Blue Origin made history when, on its first attempt, the New Shepard booster completed a suborbital flight that culminated with the autonomous landing of the booster rocket near the launch site in west Texas. The capsule landed nearby under parachutes. You can view a video of this historic flight at the following link:

https://www.youtube.com/watch?v=9pillaOxGCo

This same New Shepard booster was launched again on 22 January 2016, completed the planned suborbital flight, and again made an autonomous safe landing. This flight marked the first reuse of a booster rocket.

Again using the same hardware, New Shepard was launched on its third flight and safely recovered on 2 April 2016. On this flight, the rocket engine was re-started at a lower altitude (3,635 feet, 1,107 m) than on the previous flights to demonstrate the fast startup of the engine. The booster rocket made an on-target landing, touching down at a velocity of 4.8 mph (7.7 kph).

New Shepard landing 3Source: Blue Origin

You can view a short video of the third New Shepard flight at the following link:

https://www.blueorigin.com/news/news/pushing-the-envelope#youtubeYU3J-jKb75g

In this video, the view from the capsule at 64.6 miles (104 km) above the Earth is stunning. As the landing of the booster rocket approaches, it is dropping like a stone until the rocket engine powers up, quickly stops the descent, and brings the booster rocket in for an accurate, soft, vertical landing.

So, the current score for Blue Origin is 3 attempts and 3 successful soft, vertical landings in less than 5 months. The same New Shepard booster was used all three times (i.e., it has been reused twice).

Refer to the Blue Origin website at the following link for more information.

https://www.blueorigin.com

SpaceX Falcon 9 (F9R)

The basic flight profile for a two-stage Falcon 9 recoverable booster on an orbital mission is shown in the following diagram. For ISS re-supply missions, the target for the Dragon capsule is in a near-circular orbit at an altitude of about 250 miles (403 km) and an orbital velocity of about 17,136 mph (27,578 kph). The first stage shuts down and separates from the second stage at an altitude of about 62.1 miles (100 km) and a speed of about 4,600 mph (7,400 kph, Mach 7). These parameters are for illustrative purposes only and will vary as needed to meet the particular mission requirements. The second stage continues into orbit with a Dragon capsule or other payload.

The nine-engine first stage carries extra fuel to enable some of the booster rockets to re-start three times after stage separation to adjust trajectory, decelerate, and make a soft vertical landing on an autonomous recovery barge floating in the ocean 200 miles (320 km) or more downrange from the launch site.

The empty weight of the recoverable version of the Falcon 9 first stage (the F9R) is 56,438 pounds (25,600 kg,), which is about 5,511 pounds (2,500 kg) more than the basic, non-recoverable version (V1.1). The added fuel and structural weight to enable recovery of the first stage reduces the payload mass that can be delivered to orbit.

Falcon flight profile to barge landingSource: SpaceX

The autonomous “drone” barge is a very small target measuring about 170 ft. × 300 ft. (52 m × 91 m). It is equipped with azimuthal thrusters that provide precise positioning using GPS position data. The Falcon 9 booster knows where the drone barge should be. The Falcon 9’s four landing legs span 60 ft. (18 m), and all must land on the barge.

SpaceX_ASDSSource: SpaceX

SpaceX made a series of unsuccessful attempts to land on a drone barge before their first successful landing:

  • 10 January 2015: First attempt; hard landing; booster destroyed.
  • 11 February 2015: High seas prevented use of the barge. Instead, the Falcon 9 first stage was flown to a soft, vertical landing in the ocean, simulating a barge landing.
  • 14 April 2015: Second attempt; successful vertical landing but the booster toppled, likely due to remaining lateral momentum.
  • 7 January 2016: Third attempt; successful vertical landing but the booster toppled, likely due to a mechanical failure in one landing leg.
  • 4 March 2016: Fourth attempt, with low fuel reserve and using only three engines; hard landing; booster destroyed.

On 8 April 2016, a Falcon 9 booster was launched from Cape Canaveral on an ISS re-supply mission. The first stage of this booster rocket became the first to make a successful landing on the drone barge downrange in the Atlantic.

A002_C002_0408A9Source: SpaceX

You can view a short video of the Falcon 9 booster landing on the drone barge at the following link:

https://www.youtube.com/watch?v=RPGUQySBikQ

In the video, you will note the barge heaving in the moderate seas. After landing, the 156 foot (47.5 m) tall booster rocket is just balanced on its landing legs. Before the barge can be towed back to port, crew must board the barge and secure the booster. This is done by placing “shoes” over the landing feet and welding the shoes to the deck of the barge. Once back at Cape Canaveral, the booster will be examined and the rocket engine will be test fired to determine if the first stage can be reused.

Previously, on 21 December 2015, SpaceX successfully launched its Falcon 9 booster on an orbital mission and then landed the first stage back on the ground at Cape Canaveral. As shown in the diagram below, this involved a very different flight profile than for a Falcon 9 flight with a landing on the downrange drone barge. For the December 2015 flight, the Falcon 9 first stage had to reverse direction to fly back to Cape Canaveral from about 59 miles (95 km) downrange and then decelerate and maneuver for a soft, vertical landing about 10 minutes after launch.

Blue Origin-Falcon flight profile comparedSource: SpaceX

After recovering the booster, the Falcon 9 was inspected and the engines were successfully re-tested on 15 January 2016, on a launch pad at Cape Canaveral. I could not determine if this Falcon 9 first stage has been reused.

So, the current score for SpaceX is 6 attempts (not counting the February 2015 soft landing in the ocean) and 2 successes (one on land and one on the drone barge) in 15 months.

Refer to the SpaceX website at the following link for more information.

http://www.spacex.com

The bottom line

In the above diagram for the December 2015 Falcon 9 flight, the relative complexity of a typical New Shepard flight profile and the Falcon 9 flight profile with return to Cape Canaveral is clear. The Falcon 9 flight profile for a landing on the small, moving, down-range drone barge is even more complex.

The New Shepard sub-orbital mission is much less challenging than any Falcon 9 orbital mission. Nonetheless, both booster rockets face very similar challenges as they approach the landing site to execute an autonomous, soft, vertical landing.

Both Blue Origin and SpaceX have made tremendous technological leaps in demonstrating that a booster rocket can make an autonomous, soft, vertical landing and remain in a condition that allows its reuse in a subsequent mission. Blue Origin actually has reused their booster rocket and capsule twice, further demonstrating the maturity of reusable rocket technology.

It remains to be seen if this technology actually delivers the operating cost savings anticipated by Blue Origin and SpaceX. I hope it does. When space tourism becomes a reality, the hoped-for cost benefits of reusable booster rockets and spacecraft could affect my ticket price.

18 March 2020 Update:  Four years later

On 6 March 2020, SpaceX launched its 20th commercial resupply services mission (CRS-20) to the International Space Station (ISS).  The successful launch concluded with the 50thsuccessful landing of the first stage of a Falcon 9 launch vehicle.  On this mission, the first stage flew back for a landing at Cape Canaveral in the windiest conditions encountered to date, 25 to 30 mph.  This was the last launch with the original cargo-only version of the Dragon capsule. Subsequent launches will use 2nd-generation Dragon capsules that are roomier and designed to also accommodate astronauts.

About two weeks later, on 18 March 2020, SpaceX launched another successful Falcon 9 mission, for the first time using a first stage that had flown on four prior missions.  The satellite payload was launched into the intended orbit.  However, a malfunction in one of nine first stage engines prevented recovery of the booster rocket.

On 11 December 2019, Blue Origin reported that New Shepard mission NS-12 was successfully completed.  “This was the 6th flight for this particular New Shepard vehicle. Blue Origin has so far reused two boosters five times each consecutively, so today marks a record with this booster completing its 6th flight to space and back.”

Booster reusability has become a reality for SpaceX and Blue Origin, and other firms are following their lead by developing new reusable launch vehicles.  These are encouraging steps toward more economic access to Earth orbit and beyond.  Both SpaceX and Blue Origin have advanced reusable launch vehicle technology significantly in the past four years.  Both soon will begin human space flight using their respective launch vehicles and space capsules.

Large Autonomous Vessels will Revolutionize the U.S. Navy

Peter Lobner

In this post, I will describe two large autonomous vessels that are likely to revolutionize the way the U.S. Navy operates. The first is the Sea Hunter, originally sponsored by Defense Advanced Projects Agency (DARPA), and the second is Echo Voyager developed by Boeing.

DARPA Anti-submarine warfare (ASW) Continuous Trail Unmanned Vessel (ACTUV)

ACTUV conceptSource: DARPA

DARPA explains that the program is structured around three primary goals:

  • Demonstrate the performance potential of a surface platform conceived originally as an unmanned vessel.
    • This new design paradigm reduces constraints on conventional naval architecture elements such as layout, accessibility, crew support systems, and reserve buoyancy.
    • The objective is to produce a vessel design that exceeds state-of-the art manned vessel performance for the specified mission at a fraction of the vessel size and cost.
  •  Advance the technology for unmanned maritime system autonomous operation.
    • Enable independently deploying vessels to conduct missions spanning thousands of kilometers of range and months of duration under a sparse remote supervisory control model.
    • This includes autonomous compliance with maritime laws and conventions for safe navigation, autonomous system management for operational reliability, and autonomous interactions with an intelligent adversary.
  • Demonstrate the capability of an ACTUV vessel to use its unique sensor suite to achieve robust, continuous track of the quietest submarine targets over their entire operating envelope.

While DARPA states that ACTUV vessel is intended to detect and trail quiet diesel electric submarines, including air-independent submarines, that are rapidly proliferating among the world’s navies, that detect and track capability also should be effective against quiet nuclear submarines. The ACTUV vessel also will have capabilities to conduct counter-mine missions.

The ACTUV program is consistent with the Department of Defense (DoD) “Third Offset Strategy,” which is intended to maintain U.S. military technical supremacy over the next 20 years in the face of increasing challenges from Russia and China. An “offset strategy” identifies particular technical breakthroughs that can give the U.S. an edge over potential adversaries. In the “Third Offset Strategy”, the priority technologies include:

  • Robotics and autonomous systems: capable of assessing situations and making decisions on their own, without constant human monitoring
  • Miniaturization: enabled by taking the human being out of the weapons system
  • Big data: data fusion, with advanced, automated filtering / processing before human involvement is required.
  • Advanced manufacturing: including composite materials and additive manufacturing (3-D printing) to enable faster design / build processes and to reduce traditionally long supply chains.

You can read more about the “Third Offset Strategy” at the following link:

http://breakingdefense.com/2014/11/hagel-launches-offset-strategy-lists-key-technologies/

You also may wish to read my 19 March 2016 post on Arthur C. Clarke’s short story “Superiority.” You can decide for yourself if it relates to the “Third Offset Strategy.”

Leidos (formerly SAIC) is the prime contractor for the ACTUV technology demonstrator vessel, Sea Hunter. In August 2012, Leidos was awarded a contract valued at about $58 million to design, build, and operationally test the vessel.

In 2014, Leidos used a 32-foot (9.8 meter) surrogate vessel to demonstrate the prototype maritime autonomy system designed to control all maneuvering and mission functions of an ACTUV vessel. The first voyage of 35 nautical miles (65.8 km) was conducted in February 2014. A total of 42 days of at-sea demonstrations were conducted to validate the autonomy system.

Sea Hunter is an unarmed 145-ton full load displacement, diesel-powered, twin-screw, 132 foot (40 meters) long, trimaran that is designed to a wide range of sea conditions. It is designed to be operational up to Sea State 5 [moderate waves to 6.6 feet (2 meters) height, winds 17 – 21 knots] and to be survivable in Sea State 7 [rough weather with heavy waves up to 20 feet (6 meters) height]. The vessel is expected to have a range of about 3,850 miles (6,200 km) without maintenance or refueling and be able to deploy on missions lasting 60 – 90 days.

Sea Hunter side view cropSource: DARPA

Raytheon’s Modular Scalable Sonar System (MS3) was selected as the primary search and detection sonar for Sea Hunter. MS3 is a medium frequency sonar that is capable of active and passive search, torpedo detection and alert, and small object avoidance. In the case of Sea Hunter, the sonar array is mounted in a bulbous housing at the end of a fin that extends from the bottom of the hull; looking a bit like a modern, high-performance sailboat’s keel.

Sea Hunter will include sensor technologies to facilitate the correct identification of surface ships and other objects on the sea surface. See my 8 March 2015 post on the use of inverse synthetic aperture radar (ISAR) in such maritime surveillance applications.

During a mission, an ACTUV vessel will not be limited by its own sensor suit. The ACTUV vessel will be linked via satellite to the Navy’s worldwide data network, enabling it to be in constant contact with other resources (i.e., other ships, aircraft, and land bases) and to share data.

Sea Hunter was built at the Vigor Shipyard in Portland, Oregon. Construction price of the Sea Hunter is expected to be in the range from $22 to $23 million. The target price for subsequent vessels is $20 million.

You can view a DARPA time-lapse video of the construction and launch of Sea Hunter at the following link:

http://www.darpa.mil/attachments/ACTUVTimelapseandWalkthrough.mp4

Sea Hunter launch 1Source: DARPA

Sea Hunter lauunch 2Source: DARPA

In the above photo, you can see on the bottom of the composite hull, just forward of the propeller shafts, what appears to be a hatch. I’m just speculating, but this may be the location of a retractable sonar housing, which is shown in the first and second pictures, above.

You can get another perspective of the launch and the subsequent preliminary underway trials in the Puget Sound in the DARPA video at the following link:

http://www.darpa.mil/attachments/ACTUVTimelapseandWalkthrough.mp4

During the speed run, Sea Hunter reached a top speed of 27 knots. Following the preliminary trials, Sea Hunter was christened on 7 April 2016. Now the vessel starts an operational test phase to be conducted jointly by DARPA and the Office of Naval Research (ONR). This phase is expected to run through September 2018.

DARPA reported that it expects an ACTUV vessel to cost about $15,000 – $20,000 per day to operate. In contrast, a manned destroyer costs about $700,000 per day to operate.

The autonomous ship "Sea Hunter", developed by DARPA, is shown docked in Portland, Oregon before its christening ceremonySource: DARPA

You can find more information on the ACTUV program on the DARPA website at the following link:

http://www.darpa.mil/news-events/2016-04-07

If ACTUV is successful in demonstrating the expected search and track capabilities against quiet submarines, it will become the bane of submarine commanders anywhere in the world. Imagine the frustration of a submarine commander who is unable to break the trail of an ACTUV vessel during peacetime. During a period of conflict, an ACTUV vessel may quickly become a target for the submarine being trailed. The Navy’s future conduct of operations may depend on having lots of ACTUV vessels.

28 July 2016 update: Sea Hunter ACTUV performance testing

On 1 May 2016, Sea Hunter arrived by barge in San Diego and then started initial performance trial in local waters.

ACTUV in San Diego BaySource: U.S. Navy

You can see a video of Sea Hunter in San Diego Bay at the following link:

https://news.usni.org/2016/05/04/video-navys-unmanned-sea-hunter-arrives-in-san-diego

On 26 July 2016, Leidos reported that it had completed initial performance trials in San Diego and that the ship met or surpassed all performance objectives for speed, maneuverability, stability, seakeeping, acceleration, deceleration and fuel consumption. These tests were the first milestone in the two-year test schedule.

Leidos indicated that upcoming tests will exercise the ship’s sensors and autonomy suite with the goals of demonstrating maritime collision regulations compliance capability and proof-of-concept for different Navy missions.

4 October 2018 update:  DARPA ACTUV program completed.  Sea Hunter testing and development is being continued by the Office of Naval Research

In January 2018, DARPA completed the ACTUV program and the Sea Hunter was transferred to the Office of Naval Research (ONR), which is continuing to operate the technology demonstration vessel under its Medium Displacement Unmanned Surface Vehicle (MDUSV) program.  You can read more about the transition of the DARPA program to ONR here:
 
 
It appears that ONR is less interested in the original ACTUV mission and more interested in a general-purpose “autonomous truck” that can be configured for a variety of missions while using the basic autonomy suite demonstrated on Sea Hunter.  In December 2017, ONR awarded Leidos a contract to build the hull structure for a second autonomous vessel that is expected to be an evolutionary development of the original Sea Hunter design.  You can read more about this ONR contract award here:
 

Echo Voyager Unmanned Underwater Vehicle (UUV)

Echo Explorer - front quarter viewSource: BoeingEcho Explorer - top openSource: Boeing

Echo Voyager is the third in a family of UUVs developed by Boeing’s Phantom Works. The first two are:

  • Echo Ranger (circa 2002): 18 feet (5.5 meters) long, 5 tons displacement; maximum depth 10,000 feet; maximum mission duration about 28 hours
  • Echo Seeker (circa 2015): 32 feet (9.8 meter) long; maximum depth 20,000 feet; maximum mission duration about 3 days

Both Echo Ranger and Echo Seeker are battery powered and require a supporting surface vessel for launch and recovery at sea and for recharging the batteries. They successfully have demonstrated the ability to conduct a variety of autonomous underwater operations and to navigate safely around obstacles.

Echo Voyager, unveiled by Boeing in Huntington Beach, CA on 10 March 2016, is a much different UUV. It is designed to deploy from a pier, autonomously conduct long-duration, long-distance missions and return by itself to its departure point or some other designated destination. Development of Echo Voyager was self-funded by Boeing.

Echo Voyager is a 50-ton displacement, 51 foot (15.5 meters) long UUV that is capable of diving to a depth of 11,000 feet (3,352 meters). It has a range of about 6,500 nautical miles (12,038 km), and is expected to be capable of autonomous operations for three months or more. The vessel is designed to accommodate various “payload sections” that can extend the length of the vessel up to a maximum of 81 feet (24.7 meters).

You can view a Boeing video on the Echo Voyager at the following link:

https://www.youtube.com/watch?v=L9vPxC-qucw

The propulsion system is a hybrid diesel-electric rechargeable system. Batteries power the main electric motor, enabling a maximum speed is about 8 knots. Electrically powered auxiliary thrusters can be used to precisely position the vessel at slow speed. When the batteries require recharging,

The propulsion system is a hybrid diesel-electric rechargeable system. Batteries power the main electric motor, enabling a maximum speed is about 8 knots. Electrically powered auxiliary thrusters can be used to precisely position the vessel at slow speed. When the batteries require recharging, Echo Voyager will rise toward the surface, extend a folding mast as shown in the following pictures, and operate the diesel engine with the mast serving as a snorkel. The mast also contains sensors and antennae for communications and satellite navigation.

Echo Explorer - mast extendingSource: screenshot from Boeing video at link aboveEcho Explorer - snorkelingSource: screenshot from Boeing video at link above

The following image, also from the Boeing video, shows deployment of a payload onto the seabed.Echo Explorer - emplacing on seabedSource: screenshot from Boeing video at link above

Initial sea trials off the California coast were conducted in mid-2016.

Boeing currently does not have a military customer for Echo Voyager, but foresees the following missions as being well-suited for this type of UUV:

  • Surface and subsurface intelligence, surveillance, and reconnaissance (ISR)
  • ASW search and barrier patrol
  • Submarine decoy
  • Critical infrastructure protection
  • Mine countermeasures
  • Weapons platform

Boeing also expects civilian applications for Echo Voyager in offshore oil and gas, marine engineering, hydrography and other scientific research.

4 October 2018 update:  Progress in Echo Voyager development

Echo Voyager is based at a Boeing facility in Huntington Beach, CA.  In June 2018, Boeing reported that Echo Voyager had returned to sea for a second round of testing.  You can read more on Echo Voyager current status and the Navy’s plans for future large UUVs here:

http://www.latimes.com/business/la-fi-boeing-echo-voyager-20180623-story.html

Echo Voyager operating near the surface with mast extended. Source.  Boeing

Will Your Job Be Done By A Machine?

Peter Lobner

In September 2013, University of Oxford researchers Carl Benedikt Frey and Michael Osborne published a paper entitled, “The Future of Employment: How Susceptible are Jobs to Computerization?”. In this paper, they estimated that 47% of total U.S. jobs have a high probability of being automated and replaced by computers by 2033. Their key results are summarized in the following graphic.

Frey & Osborn key results-2013 paper

You can download their paper for free at the following link:

http://www.futuretech.ox.ac.uk/sites/futuretech.ox.ac.uk/files/The_Future_of_Employment_OMS_Working_Paper_0.pdf

On 5 Feb 2015, Fortune published an article entitled, “5 white-collar jobs robots already have taken.”  This article identifies the affected jobs as:

  • Financial and sports reporters
  • Online marketers
  • Anesthesiologists, surgeons, and diagnosticians
  • E-discovery lawyers and law firm associates
  • Financial analysts and advisors

You can read the complete article at the following link:

http://fortune.com/2015/02/25/5-jobs-that-robots-already-are-taking/

On 21 May 2015, NPR posted an interesting interactive article that provides rough estimates of the likelihood that particular jobs will become automated in the future. The ranking is based on the following factors:

  • Do you need to come up with clever solutions?
  • Are you required to personally help others?
  • Does your job require you to squeeze into small spaces?
  • Does your job require negotiation?

You can try out this interactive site at the following link:

http://www.npr.org/sections/money/2015/05/21/408234543/will-your-job-be-done-by-a-machine?utm_source=howtogeek&utm_medium=email&utm_campaign=newsletter

There is no opportunity to select many technical professions in science or engineering. Nonetheless, the results for the jobs you can select are insightful. Here are a few example screenshots from the above NPR link:

College professor automation

Aircraft mechanic automation.

Bookkeeper automation

Choosing a career is always a complicated process, but these recent studies clearly show that some careers will be marginalized by automation in the relatively near future.

First Autonomous Car to Drive (Most of the Way) Across Country

Peter Lobner

American automotive supplier Delphi modified a 2014 Audi SQ5 to make it capable of driving autonomously and then had it drive 3,400 miles on highways from San Francisco to New York City. The human “co-pilot” took control for about 1% of the distance on city streets.

image Source: www.wired.com

Read the story, including details on the car’s autonomous driving features, at the following link:

http://www.wired.com/2015/04/delphi-autonomous-car-cross-country/?utm_source=howtogeek&utm_medium=email&utm_campaign=newsletter

An important point made in this article is the great speed with which autonomous vehicle technology has advanced. In the first DARPA Grand Challenge in March 2004, all 15 competing autonomous vehicles failed to complete a very difficult 142 mile off-road course from Barstow, CA to Primm, NV. The greatest distance completed by the “winner” was 7.32 miles. In September 2005, five vehicles completed a 132 mile Grand Challenge course in southern Nevada. The third Grand Challenge in 2007 was held in an urban street environment in Victorville, CA. Six of 11 competing teams completed the course. SAIC supported a team in all three Grand Challenges.

For more information, check out the 2014 article, “The DARPA Grand Challenge – 10 Years Later,” at the following link:

http://www.darpa.mil/newsevents/releases/2014/03/13.aspx

Read details on the 2004 Grand Challenge at the following link:

http://spectrum.ieee.org/robotics/robotics-software/dusted-no-winners-in-darpas-1-million-robotic-race-across-the-mojavedesert

And details on the 2005 Grand Challenge at:

http://www.researchgate.net/profile/Erik_Blasch/publication/2961674_Unmanned_Vehicles_Come_of_Age_The_DARPA_Grand_Challenge/links/0deec525dbe44b0bea000000.pdf

And details on the 2007 urban challenge at:

http://archive.darpa.mil/grandchallenge/TechPapers/Sting_Racing.pdf

Searching the Internet of Things

Peter Lobner

The company Shodan (https://www.shodan.io) makes a search engine for Internet connected devices, which commonly is referred to as the “Internet of things”. The Shodan website explains that the intent of this search engine is to provide the following services:

Explore the Internet of Things

  • Use Shodan to discover which of your devices are connected to the Internet, where they are located, and who is using them.

Monitor Network Security

  • Keep track of all the computers on your network that are directly accessible from the Internet. Shodan lets you understand your digital footprint.

Get a Competitive Advantage

  • Who is using your product? Where are they located? Use Shodan to develop empirical market intelligence.

See the Big Picture

  • Websites are just one part of the Internet. There are power plants, smart TVs, smart appliances, and much more that can be found with Shodan.

From a security point-of-view, the last point, above, should seem a bit unsettling to the owners / operators of the power plants, smart TVs and smart appliances.

Shodan founder, John Matherly, claims to have “pinged” all devices on the internet.  Not surprisingly, the results, which are reproduced below, show that internet-connected devices are concentrated in developed nations and metropolitan areas. These results were reported on Twitter at the following link:

https://twitter.com/achillean/status/505049645245288448/photo/1

Shodan 2014 ping of Internet of Things