The intersection of technology and security has given rise to the collaboration between Everon and 1X, ushering in a new era of security solutions. This strategic partnership combines Everon's expertise in security services with 1X's innovative robotics technology to address pressing challenges in the security industry. The collaboration not only highlights the potential of security robots but also offers insights into the dynamic landscape of safeguarding businesses and properties.

Origins of the Partnership

The inception of this collaboration was spurred by a fundamental reevaluation of security needs. For Everon, the shift was inspired by the challenges of security personnel turnover and the high costs associated with guard services. Drawing from their extensive experience, Everon recognized an opportunity to leverage robotics to bridge these gaps and enhance security efficiency. This underscores the role of technology as a transformative force in shaping the security industry.

The collaboration between Everon and 1X emerged with Everonl's investment in the robotics company, cementing their commitment to advancing the realm of security through technological innovation. This partnership, which encompasses financial support and retailing of 1X's products, also involves the integration of 1X's Android EVE, referred to as Apollo within ADT, into their operational framework.

Deployment and Role of Security Androids

The Androids, known as Apollos within Everon, operate autonomously, conducting patrols and transmitting observations to Everon's Global Security Operations Center (GSOC). This integration seamlessly combines technology and human supervision, leading to a more robust security apparatus. The Androids' ability to conduct autonomous patrols and provide high-quality camera feeds for analysis exemplifies the versatility of their role.

The deployment of 1X's Androids within Everon's facilities and client settings offers a multifaceted approach to security enhancement. The Androids are not mere replacements for human security guards but rather strategic assets that augment and complement human efforts. Their primary role involves constant observation and reporting, ensuring unwavering vigilance and accurate communication of potential threats.

By enabling remote operations through VR teleoperation, the Androids offer a new dimension of flexibility. The VR teleoperation empowers remote individuals to control and interact with the Androids from any corner of the globe, fostering real-time communication and obstacle resolution.

Balancing Security and Privacy

The integration of robotics technology in security solutions naturally raises concerns about privacy and security vulnerabilities. 1X has addressed these concerns by implementing measures to ensure that Androids operate within predefined areas and adhere to privacy regulations. This approach safeguards sensitive information and fosters trust among clients and partners.

A Glimpse into the Future

The collaboration between Everon and 1X offers a glimpse into the future of security 
solutions. As technology continues to evolve, the security industry is set to witness further coordination and synergy between various systems, such as robots, drones, and analytics. 
The capabilities of these systems hold the potential to thwart threats before they materialize, minimizing the need for human intervention.

1-minute summary:

• This blog post discusses how 1X collects data for training our androids
• We believe that data quality > data quantity > algorithms
• Our data collection team can fine-tune models themselves to customize behaviors

1X’s mission is to create androids that work alongside people, and use them to meet the world's labor demands for an abundant society. 

Traditional robotics and industrial automation solutions already make society very productive. Wonders of automation can convert 10 tons of potatoes into potato chips in a few minutes, assemble about a billion smartphones per year, and manufacture a car from scratch every 60 seconds. In such factories, the repetitive task is done in high volumes so this warrants the effort to build a custom machine.

Learning to do chores in ever-changing environments

On the other hand, there is a vast “long tail” of chores for helping humans in human spaces: keeping the office safe and secure, carrying groceries from the car, sorting trash for recycling, tidying and cleaning indoor environments, removing debris and litter from public areas. Existing robot products have yet to make a big dent on these tasks. To tackle these, we have to create a general-purpose robot with the same physical affordances as a human (i.e. an Android), and they have to be smart enough to do everyday chores in the office or the home.

Home and office environments are challenging because they are unstructured and always changing from human use. For example, at one of our customer sites where we have deployed our patrolling solution, there is such frequent construction activity that the location of obstacles and barriers are changing from day-to-day. Because of this, our patrol solution cannot assume that the locations and appearances of obstacles remain constant. Generally, automation becomes challenging when the software developer can’t assume much about the state of the world outside of the robot’s body. You cannot assume where objects are placed relative to the robot’s grippers, or whether the desk in a map has been moved, or whether the coffee tin has enough coffee left in it to brew a cup.

Defining Behavior Through Data

Our own approach to autonomy is inspired by how digital assistants like ChatGPT and autonomous vehicles (Waymo) are developed. The strategy is to collect a large variety of environmental situations the droid encounters in data, and learn a general understanding of the task from that data, rather than hand-engineering the code to perform a single repetitive motion. By collecting large amounts of diverse experience, our droids generalize to new situations they have not seen before. Initially they don’t know what concepts like “grasping” or “sorting” or “patrolling” mean, but when provided with enough examples of these tasks across a wide variety of scenarios, they develop a general understanding of what to do in new environments.

Because our androids’ understanding of the world is derived from data, how we collect and curate our training data becomes a critical part of our strategy. Here’s what matters the most to us when collecting data:

Embodied teleoperation in VR tells us how hard it is to learn tasks:

If a human can look through the android’s eyes and control it to perform the task using VR teleoperation, then in principle it should be possible to replicate the human decisions with a neural network to perform the same task with the same inputs. When exploring a new task we want to teach the robot, we first verify that it is feasible in VR. This is the existence proof that there exists at least one neural network (the human brain) that can perform the task with the information available to the droid’s sensors.

Our VR data collection system also gives us an intuitive guess of the difficulty of learning the task. All other things being equal, predicting the actions for a 2 second demonstration of opening a door is much easier than predicting the actions for a 20 second demonstration of opening the same door. Machine learning methods for robotics tend to have an easier time predicting a short sequence of clean actions than a long sequence of noisy actions. Any extra unnecessary time spent performing the demo effectively becomes “noisy data” that adds to the difficulty of training our models. To train as efficiently as possible, we care deeply about making the most intuitive, low-latency teleoperation interface possible.

Optimizing the data collection tools to be easy to use directly translates to cleaner, shorter data and more capable androids.

‍

Investing in high quality data to train good models 

It is an open secret in the applied ML community that when it comes to training performant ML systems, whether it be physical robots or digital assistants, the careful curation of training data is often far more impactful than developing new learning algorithms. By selectively gathering labels for scenarios where the model fails, and then re-training the model on that new data, we can fix the failure modes without changing the underlying algorithm. 

Our engineers who train our ML models spend significant amounts of time practicing tasks in VR and reviewing the data to ensure that the way we gather and process data is as time-efficient as possible. 

We also employ a team of Android Operators to scale up data collection to more diverse environments. If having a detailed understanding of the data being collected allows a ML researcher to more effectively train a good model, then the converse is also true: the person responsible for collecting data for training models can become more effective if they train some models themselves. They can build a detailed intuition for how much behavior change they can expect from the model as they vary the quantity and quality of data they collect. 

Open source GUI wrappers around stable diffusion have allowed non-ML experts to fine-tune the base stable diffusion models to add on new styles and improvements. Inspired by this trend, we’ve built similar tools that allow our android operations team to fine-tune behaviors. 

The AI team trains the base model, which has a general visual understanding of the world. The Android Operations team designs new tasks, collects data, trains and deploys the models, and collects more data in situations where the model struggles to generalize. Here are a few of the behaviors that our Android Operations team have taught the robot on their own:

‍

‍

‍

‍

As robotic capabilities become more and more data-driven over time and less dependent on specialist knowledge,robotics will become more accessible to non-technical users.

On November 4th 2023 1X is hosting our first AI event, 1X Discover Day: Embodied Learning opening up limited invitations only. Click to learn more.

Robots have been a symbol of technological progress and innovation for decades. However, the challenge of replicating human dexterity and efficiency in robots has been formidable. Central to this challenge is the use of gears in most robotic systems. While gears provide power, they can also lead to added weight, reduce natural dynamics, and hinder agility. At 1X, we're taking an innovative approach to bridge this gap.

So, why don’t all roboticists forgo the gear system? Historically, it has been a tall order to develop motors that deliver sufficient force and torque without relying on gears. But as the demand for more efficient, agile, and human-like robots grows, so does the need for advancements in motor technology. At 1X, we’ve done just that. We've successfully developed motors that achieve about 80% of the force density of human muscle without the use of gears. This technological leap surpasses the current world record by approximately five and a half times, paving the way for a bio-inspired approach to robotics.

Such advancements are more than just technical milestones.

They represent a shift in how robots can be integrated into society. By achieving a more natural motion, robots can better adapt to their surroundings, resulting in safer interactions with humans and environments. This becomes crucial when we consider their potential applications in real-world settings.

For a visual analogy, consider the inertia in our limbs when we move. A humanoid robot, using traditional technology with a gear ratio of 80:1, might experience an inertia equivalent to carrying a 24-kilogram weight on its foot. It's no wonder then that these robots often appear stiff and mechanical in their movements. The 1X system has dramatically reduced this inertia, equivalent to a mere 0.5 kilograms for the feet. Such reductions are not just impressive on paper; they translate to significant enhancements in robot agility, efficiency, 
and safety.

Large language models, like ChatGPT and multimodal models such as GPT-4, represent a significant leap in understanding and interpreting the world around them. However, for machines to reach a true level of general intelligence, they need the capability to physically interact with their surroundings. This interaction offers a unique dataset, allowing robots to understand and adapt to various modalities and behaviors.

However, scaling this interaction presents challenges. Using ChatGPT as an example, which is trained on three trillion tokens, adapting a similar scale to robot movements means potentially needing millions of robots operating round the clock. This monumental task underscores the significance of physical labor in the robotics equation.

To address this, 1X is heavily investing in shared autonomy. This involves a human-robot collaboration, where robots can request human intervention in complex tasks. Our Android, EVE, for example, can patrol a facility autonomously but might need human assistance for unexpected challenges, like an obstructed door. By integrating shared autonomy, we not only provide immediate utility to customers but also create a valuable data feedback loop, helping our robots continuously learn and adapt.

Our next-generation android, NEO, encapsulates our vision for the future.

While maintaining a lightweight design, NEO can replicate human speeds and even surpass human strength. By focusing on passive safety, NEO can interact with its surroundings and people without causing harm, allowing it to explore and learn autonomously.

In conclusion, predicting the arrival of truly autonomous, human-like robots is challenging. Whether it takes 10 years or 30. My hope is that these innovations will lead to a sustainable abundance, where Androids, powered by renewable energy, can produce goods and services at negligible costs and minimal environmental impact.

1X will be attending IEEE's Humanoids Conference in Austin, Texas this year as a platinum partner. Our sponsorship represents 1X’s commitment to further the development, access to the top talent and expand our understanding of humanoid technology.

The IEEE's Humanoids Conference, scheduled from the 12th to the 14th of December, is the premier annual gathering for researchers, academics, and industry professionals dedicated to the study of humanoid robots. Participants will get to experience a range of inspiring talks, interactive workshops, and groundbreaking exhibits.

As a sponsor, 1X will also have an exhibit allowing the community to come meet the crew, hardware and talk about our latest innovations. Additionally, a representative from 1X will be taking the stage to discuss our advancements in the field of humanoids.

To know more about the conference or to register:


IEE's Humanoids Conference Official Page
Conference Program

We hope to see many of you there and look forward to sharing our passion for humanoid technology with the community!