Kawasaki Robotics //treeforeurope.com/in/ Wed, 25 Mar 2026 00:50:40 +0000 en-AU hourly 1 //www.altis-dxp.com/?v=6.8.4 //treeforeurope.com/tachyon/sites/37/2022/02/cropped-site-icon.png?fit=32%2C32 Kawasaki Robotics //treeforeurope.com/in/ 32 32 Kawasaki Robotics //treeforeurope.com/in/blog/20260316_kaleido9/ Wed, 25 Mar 2026 00:49:25 +0000 urn:uuid:5cdb2b8f-3f06-4cc5-abce-3190864cb589 In recent years, interest in humanoid robots has been growing rapidly around the world.
Kawasaki Heavy Industries, however, began its research and development of humanoid robots back in 2015—well before the current surge of attention. Nearly a decade has passed since then, marked by continuous improvements and steady progress. As a result, Kaleido 9 has emerged as the ninth generation in this ongoing evolution.

Humanoids Are Not All-Purpose. That’s Exactly Why Their Role Is Now Clear

Dedicated machines and industrial robots are highly rational solutions for processes with predefined tasks and production lines that require strict takt time control.
Even so, there are situations where the use of a humanoid robot makes clear sense.

For example:
  • Equipment, tools, and even stairs designed for humans can be used as they are
  • There is no need to prepare or redesign the environment specifically for robots
  • Humanoids can operate flexibly in situations where tasks are difficult to predict in advance, such as disaster response sites
We believe that the greatest value of humanoid robots lies in the concept of “adapting robots to the human world,?rather than forcing humans to adapt their environments to robots.

Key Evolutionary Points of Kaleido 9

01. Structural Design That Dramatically Enhances Fall Resistance

Kaleido 9 features significant structural enhancements compared with the previous generation, Kaleido 8.
With commercial deployment in mind, the leg structure has been redesigned to strictly comply with component ratings. By increasing movement speed while maintaining torque, the legs can now respond more quickly when the robot loses its balance.

This ability to step out rapidly and regain stability has dramatically improved overall robustness—an attribute that Kawasaki Heavy Industries considers fundamental to the design of its humanoid robots.

02. Autonomous Walking That Sees, Judges, and Walks

Kaleido9 is capable of autonomous walking while continuously recognizing and understanding its surroundings. In the demonstration, Kaleido 9 was shown making human-like decisions in real time—detecting and avoiding people and obstacles, recognizing steps on stairways, and automatically adjusting its stride to avoid stepping on edges and corners. This autonomous walking capability is supported by LiDAR (Light Detection and Ranging)-based self-localization using SLAM (Simultaneous Localization and Mapping).
Autonomous navigation with obstacle avoidance

03. An Intuitive Remote Control Solution Designed for Real-World Applications

Fully autonomous humanoids equipped with AI still face many technological challenges. Rather than pursuing full autonomy all at once, Kawasaki Heavy Industries envisions a set of realistic, step-by-step approaches. These include:
  • Remote-controlled operation, allowing humans to directly intervene when needed
  • Hybrid operation, where humans handle high-level tasks while routine or simple tasks are delegated to AI
Kaleido 9 supports intuitive, real-time remote operation through a head-mounted display and has already been demonstrated in disaster response scenarios, highlighting its potential as a practical real-world solution.
Remote-controlled disaster rescue demonstration

More Than Just Walking: The Idea Behind the Kaleido Station

Another distinctive feature of Kaleido 9 is the Kaleido Station.
This mobile platform is designed to enable efficient long-distance travel on level ground using wheels, while Kaleido itself stands up and performs tasks once it reaches the work site.

Looking ahead, the Kaleido Station is also planned to function as a charging station. By combining the strengths of both humanoid and wheeled mobility, Kawasaki Heavy Industries is pursuing a realistic and practical approach to social implementation.

A Technology Roadmap for Humanoids: Evolving Step by Step

Humanoid robots will not suddenly become capable of doing everything autonomously.
Kawasaki believes that the implementation of humanoids in society will progress in stages.

Around 2030: Independent Operation and Support in Factories and Other Environments

Around 2030, humanoid robots are expected to be used primarily in controlled environments such as factories and plants. In these settings, they will perform independent work under remote control and support people in their daily operations. At this stage, the key technologies required will be the fundamental elements that connect humans and robots. These include remote operation, conversational interaction, voice-based instructions, SLAM for autonomous mobility, and environmental awareness. One of the major advantages of humanoid robots is their ability to use equipment and tools originally designed for humans without modification. In this model, humans remain responsible for decision-making, while robots leverage their physical capabilities to carry out tasks. Main application scenarios include:
  • Communication support in medical and nursing care settings
  • Simple tasks such as transportation and patrol duties
Key technologies and capabilities required:
  • Environmental awareness
  • Remote operation
  • Conversational interaction and response to voice instructions
  • SLAM-based autonomous mobility

Around 2040: Autonomous Operation in Unstructured Environments and a Broader Range of Applications

Looking ahead to around 2040, humanoid robots are expected to be capable of operating autonomously in unstructured environments while continuously recognizing and understanding their surroundings. Their applications are envisioned to expand into more complex and highly dynamic workplaces, including tasks performed at heights, in confined spaces, and at disaster response sites. This phase will require a wide range of advanced technologies, such as enhanced environmental awareness, posture control, contact sensing, dexterous hand control, object recognition, and swarm control. In addition to core capabilities like autonomous mobility and environmental perception, advanced control technologies that directly enable task execution will play a key role in supporting these applications. Main application scenarios include:
  • Spatial mobility in challenging environments, such as work at heights and in confined spaces
  • High-performance tasks, including assembly and maintenance
Key technologies and capabilities required:
  • Response to voice instructions
  • Contact sensors
  • Environmental awareness
  • Posture control
  • Hand control
  • Swarm control
  • Object recognition

Around 2050: Toward Humanoids for All Environments and Tasks

By around 2050, the goal is for humanoid robots to evolve into systems capable of responding flexibly to all environments and tasks, both indoors and outdoors. At this stage, in addition to technologies that enable operation across diverse environments, a high level of environmental resistance will be essential. This includes resistance to water and dust, as well as tolerance to heat, cold, radiation, and corrosive gases.

The essential value of humanoid robots lies in their ability to move and operate in the same environments as humans, carrying out tasks on their behalf—even in situations where the nature of the work cannot be clearly defined in advance. Main application scenarios include:
  • Operation in all environments, including outdoor and disaster response settings
Key technologies and performance required:
  • Environmental resistance, such as waterproofing, dustproofing, heat resistance, cold resistance, radiation resistance, and resistance to corrosive gases

The Ultimate Goal: Disaster Response Sites

The ultimate destination for Kawasaki’s humanoid development is disaster sites. Japan is prone to natural disasters, and the need to respond to situations that cannot be foreseen in advance. Kaleido 9 is a solid step towards the future.

Related blog

Humanoid Robot: The Evolution of Kawasaki’s Challenge
]]> Kawasaki Robotics //treeforeurope.com/in/blog/20260226_physicalai/ Thu, 05 Mar 2026 06:39:17 +0000 urn:uuid:92498198-ec81-4b72-8524-f870d4f8d85d
Advances in AI are dramatically expanding the possibilities of robotics.
One of the key concepts gaining attention in recent years is “Physical AI.?br>Physical AI refers to AI that is not only intelligent, but also capable of moving, working, and interacting with people in the real world.
The true value of next-generation robots lies in how effectively they can be deployed and contribute to society.

What is physical AI – from ‘thinking AI’ to ‘acting AI’?

Until now, AI has primarily focused on decision-making and optimization in the digital world.
Physical AI, by contrast, refers to AI that acts in the real world—using sensors and physical bodies such as robots to move and manipulate objects.

The real world is inherently unpredictable, with constantly changing environments. In such conditions, what matters most is not perfect judgment, but the ability to operate safely, reliably, and consistently.
The most critical requirement for robots is that they never fail. More than achieving success once, the true value of physical AI is defined by its ability to deliver stable, continuous performance over time.

Classification.Corresponds to
Humanfunctions		 Main role Specific examples
Perception AI Eyes & Ears Environmental awareness, image and sound recognition Camera recognition for automated driving,
Voice assistants
Generative AIBrain Information generation, creation Image generation AI
Agentic AIIntention Planning, decision-making, autonomous action Autonomous agents,
Business automation
Physical AI Physical Physical movement, control Industrial robots, delivery robots
Four categories and features of AI
Vision and AI enable robots to perceive their environment and act accordingly.

Humanoid robots are not “finished??the value of Physical AI continues to expand, step by step.

Humanoid robots have attracted attention as a symbol of Physical AI. Since 2015, Kawasaki has been steadily developing its humanoid robot, Kaleido. However, we do not view humanoids as entities that will suddenly transform society. The widespread adoption of humanoid robots will begin by supporting and complementing human tasks. From there, they will be applied to specific tasks and environments, gradually expanding their roles over time. We believe that this step-by-step social implementation is the most realistic approach. The key is not to replace all human work, but to design a division of labor in which humans and robots can each make the most of their respective strengths. This is where the true path to implementing Physical AI lies.

Kawasaki Robotics?strength lies in its on-site capability.

Kawasaki Robotics places great importance on robots being used continuously in real-world environments. In manufacturing and other demanding workplaces, robots are expected to operate safely, reliably, and without interruption—while maintaining consistent quality. Over many years, we have worked closely with these sites to refine machinery, control systems, safety, and operation as an integrated whole. This field-oriented approach to manufacturing is one of Kawasaki Robotics?core strengths, and we see it as the foundation of our competitiveness in the age of Physical AI. Rather than following the evolution of AI as a mere technological trend, Kawasaki Robotics promotes the implementation of Physical AI from the perspective of how it can directly contribute to solving real-world challenges. Our goal is to create environments where humans and robots can collaborate effectively, leveraging their respective strengths—step by step, at each individual site. Through the steady accumulation of these efforts, robots will evolve into an indispensable part of society. This is the future of Physical AI, and the next stage of robotics as envisioned by Kawasaki Heavy Industries. ]]> Kawasaki Robotics //treeforeurope.com/in/blog/202511_kaleido/ Thu, 15 Jan 2026 08:28:27 +0000 urn:uuid:d4d42d01-2b1c-49ab-8cae-8b54da53184e Kawasaki Robotics, driven by its purpose “Unlocking human ingenuity to create robotics that enrich the future?/em> has long pursued the development of technologies that enable robots and people to coexist. The ultimate form of this coexistence is the humanoid robot—capable of moving on two legs and performing tasks with two arms, giving it the versatility necessary to take on work traditionally done by humans.
This article introduces the lineage of Kawasaki’s humanoid robots, from the first prototype to the eighth generation.

Origins of Kawasaki’s Humanoid Development

Kawasaki, the company that built Japan’s first industrial robot, has long led the field of industrial robotics for factory production lines. Drawing on more than half a century of experience in developing and manufacturing industrial robots, Kawasaki has also expanded into human‑centric robotics—creating systems that provide value outside traditional factory cells. These include the duAro dual-arm collaborative robot, designed to work safely beside humans, and Successor, a remote‑cooperation system that removes the difficulty of conventional robot teaching. As the ultimate extension of these efforts, Kawasaki is developing humanoid robots. Because humanoids walk on two legs and perform tasks with two arms, they offer a high degree of freedom and the potential to take on work traditionally performed by humans.
The Evolution of the Kaleido Series

2015 ?The First Prototype

Kawasaki began developing humanoid robots around 2015. For a company that had long specialized in industrial robots, creating a robot capable of bipedal locomotion represented a major challenge. To determine how to reproduce a human‑like form mechanically, the team first explored how motors and joint mechanisms should be arranged. However, the deeper we studied human anatomy, the more we realized just how complex and flexible the human body truly is. This led to a shift in design philosophy: instead of trying to replicate human motion directly, we extracted only the essential functions of the human body and recreated them using motors and mechanical structures. This conceptual shift ultimately enabled the realization of stable bipedal walking. One of the biggest differences between industrial robots and humanoid robots is the load they must support. A typical industrial robot with a 100‑kg payload often weighs more than 600 kg and is designed to lift objects lighter than itself. In contrast, a humanoid must lift loads greater than its own body weight, making extreme weight reduction critically important.
The first prototype had insufficient leg rigidity, and its knee mechanism in particular was vulnerable to torsion, resulting in unstable walking.
Kaleido1号機
Up to the third prototype, those were used the E‑controller designed for industrial robots. However, it was large, had a control cycle too slow for humanoid applications, and required four separate units, which meant the controllers alone occupied a significant amount of space. These issues were later resolved by transitioning first to the F‑controller and eventually to a dedicated humanoid controller.
Falling poses the greatest risk to a humanoid robot, as it can lead to serious damage to the machine. To ensure the robot could withstand such incidents, we conducted repeated tests in which the robot was intentionally made to fall while performing controlled break‑fall motions.

2017 ?First Public Debut at iREX

Kawasaki publicly unveiled Kaleido at the 2017 International Robot Exhibition (iREX 2017), demonstrating actions such as stand‑up motions and pull‑ups.

  • Height: 175 cm
  • Weight: 85 kg
  • Power: External tethered supply
国際ロボット?017の様? class=
Behind the scenes:
The robot suffered issues until the eveniBehind the scenes:
The robot suffered issues until the evening before the exhibition, forcing the team to consider reducing the demo drastically.
A final round of adjustments worked, enabling a successful full demonstration on opening day.
FコントローラーとKaleido

Although switching to the F‑controller made the system somewhat more compact,
it still weighed about 30 kilograms—far too heavy to be mounted inside a humanoid robot.

2019 ?Battery Integration and the Path to Lightweight Design

Kawasaki unveiled its first fully untethered bipedal walking demo with an onboard battery. At that time, Kaleido stood 178 cm tall and weighed 85 kg. While maintaining this weight, the team replaced the four external controllers with compact, high‑performance amplifiers and motor drivers, and integrated all electrical components into the robot’s body. By adopting an onboard battery as the power source, Kaleido achieved complete stand‑alone operation. To realize stand‑alone operation without increasing the overall weight, numerous development efforts were undertaken. The first focus was the skeleton and exterior design. Magnesium alloy—lighter than aluminum—was used for structural components. Additionally, the metal‑frame exterior panels were replaced with 3D‑printed resin parts, resulting in significant weight reduction. Producing resin components in‑house also dramatically accelerated the prototyping cycle. Beyond these major improvements, the team performed countless incremental weight‑saving refinements. At times, components became too fragile after being lightened, forcing the team to revert them to stronger, heavier versions. It was extremely difficult to predict where issues would arise, so the team repeatedly built, tested, and refined the design through persistent trial and error.

Walking Control Development and Custom Force Sensors

Once hardware reached maturity, the team began developing advanced gait control. Humanoids typically rely on 6‑axis force/torque sensors at the ankles, but commercial sensors were:

  • Highly accurate
  • Too heavy
  • Expensive
  • Susceptible to shock damage

For a humanoid performing steps, stomps, or jumps, lighter and shock‑resistant sensors were a necessity. Kawasaki therefore developed custom 6‑axis force/torque sensors optimized for humanoid walking.

Kaleidoの力センサ開? class=

2021 ?Introduction of the “Friends?/strong>

Compared to the powerful and robust Kaleido, the Friends series focused on a slimmer, more approachable humanoid designed for tight spaces and potential future roles in caregiving and daily-life assistance.

A Friendly, Approachable Design

Friends was designed with soft, curved lines so it can blend naturally into human environments such as homes and care facilities. To create a friendly impression, the robot uses a display for its eyes, allowing it to express emotions in a human‑like way. The name “Friends?reflects the idea behind the project—a robot that can truly become a friend to people.

Conversational and Gesture Interaction

In collaboration with Osaka University’s Nagai Laboratory, we developed conversational and gesture‑based interaction capabilities. This was the first attempt of its kind among the humanoid robots we had created. Using AI, Friends generates spoken responses to questions posed by users. At the same time, it performs gestures synchronized with its replies, enabling more human‑like communication.
ヒューマノイドロボットFriends
At an event held at the National Museum of Emerging Science and Innovation, Friends answered questions from children in the audience. Although the conversational quality was not perfect, the venue was packed, and the robot’s humorous interactions often drew laughter from the crowd.

2023 ?Achieving Stable Walking

Kawasaki upgraded both software and hardware, adding real-time footstep adjustment to correct landing positions when balance is disturbed. This significantly reduced the risk of falling and improved robustness.

The Future of Humanoid Robot Development

Robots are expected to help people by taking on work that is dangerous, dirty, or physically demanding. However, robots are traditionally good at repeating tasks that have been pre‑taught, which makes it challenging for them to handle work that cannot be predicted in advance. If humanoid robots can be operated remotely, they could take over hazardous tasks at dangerous sites, while operators remain safely in urban areas and control them from anywhere in the world. This capability is especially valuable in disaster‑response scenarios, where unstable terrain and the risk of secondary accidents make human entry extremely dangerous. Kawasaki’s booth is dedicated to realizing a future where humanoid robots work right alongside you.

]]> Kawasaki Robotics //treeforeurope.com/asia-oceania/case-studies/case_yamaguchi-mfg/ Mon, 25 Aug 2025 06:15:52 +0000 urn:uuid:5e00b811-0bc2-4647-8bc5-b871b6779d94 Kawasaki Robotics //treeforeurope.com/asia-oceania/case-studies/pharmaceutical-automation-with-kawasaki-robotics-safe-and-reliable-production-of-anticancer-drugs/ Mon, 04 Aug 2025 00:28:21 +0000 urn:uuid:72fca0c8-33ae-4302-85cf-92ad815adb99 Kawasaki Robotics //treeforeurope.com/in/blog/ready-for-a-robot-integration-insights-for-welding-applications/ Wed, 12 Mar 2025 06:46:04 +0000 urn:uuid:4e0be900-9038-499a-81c8-3706aa4686d4 By Brandon Day, Senior Engineer Robotic Metal Fabrication

Robot integration in many welding applications makes sense, given that welding is not conducive to pleasant working conditions for humans. Not only is it a hot and dirty process, but it also emits unhealthy chemicals such as hexavalent chromium when welding stainless steels and other chromium-containing material. Although these health concerns might seem like enough reason to implement a robot, there are other considerations.

To ensure that robot automation is beneficial for an application, Kawasaki Robotics North America recommends evaluating these four questions for the best return on investment:

1. What is the part size?

Small parts that fit inside the welding cell are ideal for robot welding, but larger sized parts are possible.

2. How large is the volume size?

Best case scenario for robot welding is high volume (thousands of parts per day, hundreds of thousands per year). Also, consider whether the application includes volumes high enough to create the return that is necessary to pay off the machine in a respectable amount of time.

3. Are tolerances tight enough?

Tight tolerances on parts are necessary to provide accurate repeatability. The robot welds on the exact same spot every time; therefore, too much tolerance will cause the weld to miss often.
Kawasaki robot spot welding sedan

4. Is manpower an issue?

A worker shortage might demand robot integration in a welding application.

Although robots offer accuracy, repeatability and labor and cost savings, human welders are still a necessity. Their knowledge of the art ?including the ins and outs such as torch angles, gasses, metrology, metallurgy ?is required to properly program robots for these applications. Because every major manufacturer has its own welding specifications, it is key for every facility
that performs welding to have a specialist on staff who can interpret these different specs.

Robot Programming Made Easy

When robots are added to an application, instead of doing the welding themselves, welders can learn the job of programming the robots. Progressing into this higher skillset leads to not only a higher paying position where a welder is running multiple robot cells, but also a healthier work environment.

Kawasaki Robotics North America makes it easy for anyone with basic welding knowledge, including engineers, technicians and programmers, to learn robot welding programming quickly and thoroughly. The robot provider offers a three-and-a-half-day 168ѯ¼:training course, “Arc Operations and Programming,?at its training center in Wixom, Michigan.

The class is designed for basic robot and arc welding operations, programming and safety training. It includes topics such as operating controls, selecting proper menus for programming, positioning the robot by use of pendant control, and program creation procedures and modification techniques. Upon course completion, students will know how to run the robot system in both manual and automatic modes, create a block step program using weld conditions, teach a path using weld and non-weld steps, and modify programs.

]]> Kawasaki Robotics //treeforeurope.com/in/blog/202404_ai/ Wed, 26 Feb 2025 05:41:48 +0000 urn:uuid:97f35181-2749-4da6-a678-c8b0865f0b57

Kippei Matsuda: Robot Technology Development Department, System Technology Development Center, Technology Development Division. Dr. (Engineering)

Describe the NFL challenge, the solution, and what made it exciting?

The competition aimed to identify player collisions, with more than 1,000 data scientists worldwide competing to build an accurate solution using NFL game footage and sensor information. American Football is known to be one of the toughest, physically demanding sports in the world involving high-impact contact, and although players wear protective gear, injuries are prevalent. Head collisions, in particular, often result in serious injuries or disabilities, and it’s been challenging to find ways to reduce the impact of collisions along with effective treatment after an injury occurs. If we could accurately identify which players suffered head impacts during a match, we could effectively administer treatment and advance the research into the effects of helmets and how to mitigate the impact. Doing this type of research manually would be extremely time-consuming, so the NFL held this competition to use AI technology to solve this problem.

The key to winning this competition was the successful integration of two distinct data types, video and sensor information. We utilized video images to identify player collisions and estimated player positions using sensors attached to the players. The development of an AI system that meticulously analyzed and predicted even the smallest changes in player position and posture, such as crouching or falling, set us apart. This innovation significantly improved our accuracy compared to other participants. As a result, our processing speed was 83 times faster than manual operation, and tasks that took 3-4 days could now be completed in just 2 hours, which was greatly appreciated.

Kippei Matsuda
In general, AI image analysis involves detecting objects in images, but in this competition, we had to consider the three-dimensional (3D) positions of the players on the field on the screen, which was an exciting challenge that we had never done before. It is difficult to analyze data because you have to look at it over and over again, but the NFL video was so powerful that I enjoyed watching it over and over again. Thanks to this, by the end of the competition, I was able to imagine the players’ movements just by looking at the titles of the videos.

What made you participate in the NFL challenge & what were the results?

It all started with me thinking it could be part of my studies. Theoretical aspects can be learned from books and other sources, but knowing how to use actual data and run a simulation is difficult. As a developer, touching and analyzing data and repeating trial and error is essential. I was attracted by the competition as it provided materials that led to practical learning.

“Participating in the competition allowed me to put AI development into practice”

Kippei Matsuda
Honestly, I never felt the competition itself was hard; what was difficult was finding time for my studies and family, as worked on this during my personal time. When I was playing with my children at the park, I would suddenly think, “Maybe I could do that part this way,” and it would bother me. It was hard to relax. I was on the train, and I was shaking. Since the start of this project, I did not think I could win, but when I did, I was thrilled. I have no complaints about winning and was happy to share the news with everyone around me. I could hardly get any work done that day!

How do you work with AI vision today at Kawasaki Robotics?

I am currently involved in developing products that utilize AI vision specifically for robotics. AI analyzes images captured by cameras and processes them in various ways; for example, in our depalletization solution, it processes images of the product that need to be unloaded. Depalletizing solutions are used to improve unloading efficiency at distribution centers and factories. The Depalletizing Solution is equipped with 3D AI vision and is capable of highly sophisticated analysis of the cargo it handles.
Kawasaki Robotics depalletizing solution is equipped with 3D AI vision and is capable of highly sophisticated analysis of the cargo it handles. By specializing in unloading, we have achieved high performance at a low cost.Kippei Matsuda
Manual unloading is very costly and time-consuming, and conventional robotic solutions are not flexible enough to handle the work. In this respect, our depalletizing solution has succeeded in increasing the accuracy and speed of automated unloading operations compared to conventional solutions.
For example, with conventional robotic systems solutions, all package sizes and shapes must be registered before being picked. If packages of unregistered shapes make it into the workflow, they cannot be processed. Our depalletizing solution however requires only the smallest and largest sizes to be registered, and all packages can be processed. In conventional robotic systems the robot needs to know the correct size and shape of the product being handled. A camera is required in order to recognize and confirm the product size and shape. If it’s confirmed as correct the robot arm will pick the product, to teach the system this is a time-consuming process.
Our depalletizing solution requires no prior product registration other than the minimum and maximum dimensions, significantly reducing teaching time.Kippei Matsuda
The main reason why Kawasaki Robotics’ depalletization solution has been successful is that we pair low-cost camera hardware with our robust AI software to process complex product images without a bunch of ad-ons. We achieved a high-performing and easy-to-use solution by focusing on developing a depalletizing system, and by narrowing down the functions, we achieved better cost performance than other companies’ products, making it easier for companies to introduce the product.
Mr. Himekawa, leader of the product development (belongs to General-Purpose System Section 2, General-Purpose System Department, Robot Division)

How will depalletizing and AI solutions continue to develop?

I believe that the efficiency of depalletizing solutions will increase as more and more data is collected and AI learning progresses. For example, one of the difficulties in developing a depalletizing solution was the strings and tapes on the surface of packages. When AI sees them, it may mistake the strings for the boundaries of the cardboard. It may then decide that the box is smaller than it actually is and take it by mistake. However, if the system is used at various sites in the future and data is collected, it will be able to learn multiple variations of packages. Then, even if a box has strings, tape, or stickers, the robot can unload it appropriately based on its past experience without being misled. Collecting a lot of good-quality data for the robot will be necessary. I also think that AI, called the infrastructure model, will be key. One example is ChatGPT, which has been attracting much attention in recent years. Since the underlying model is trained based on a large amount of data, it has a very high recognition capability, a kind of common sense. Fundamental models can handle a variety of information, including text, images, and sound, and have the potential to dramatically expand the use of robots, not only in logistics. In the future, I would like to expand the use of AI by making good use of data and the basic model.

Click here for Depalletizing Solution product page(Japanese)

]]> Kawasaki Robotics //treeforeurope.com/in/blog/story_21/ Thu, 09 Mar 2023 00:51:44 +0000 urn:uuid:35c957ca-e3e9-4766-837f-d29a46356b05 Human coexistence type humanoid robot RHP Friends (hereinafter called “Friends?
ー　Tell us the background leading to the development of RHP Friends.
At Kawasaki, we are working with the National Institute of Advanced Industrial Science and Technology on joint development for the improvement of the humanoid robot technology in Japan. What kind of thing is a humanoid robot that will be useful for the Japanese industry? We had assumed disaster rescue as one example for the implementation of Kaleido in society, but implementation in factories could be even quicker. Even within a factory, we are not talking about the manufacturing of mass produced items, but rather works such as the assembly of products with high added value. Wouldn’t it be possible to use humanoid robots for this work? In workplaces such as that, there are often humans bending their bodies flexibly in tight spaces to perform the tasks. It is also sometimes the case that it is necessary to travel along narrow passageways just to get to the site of the work. However, the work itself is only plain and easy tasks such as tightening screws. In such cases, robots can do better than humans. However, to be able to perform work in such places, a humanoid robot would have to be a slender model. It would have to be flexible and designed to be safe for working together with humans, and it must not damage the surrounding equipment even if it fell over. It is always good to have strength, but this work does not actually require very much strength. There must surely also be opportunities for a robot such as this to be useful for aircraft engines, gas turbines and places such as chemical plants. What we were previously aiming for with our development of Kaleido was robustness and power. In response to this, the origin of the Friends project was when we decided to develop a robot that would have a slim body that could travel along narrow paths, and also a design that the people around would consider it to be gentle.

ー　Tell us the points that have been passed on from Kaleido to Friends, and also what points are different.
Know-how and design concepts from 50 years of industrial robots at Kawasaki have been passed on, not only to Kaleido, but also to Friends. We have also followed the designing technology for Kaleido. For the points of difference, the first is the slimmer profile. To achieve this, we use high-quality and compact motors to reduce the weight. Kaleido weighs around 80 kg, but Friends is only around 55 kg. Furthermore, the compact motors make installation in the slim body of Friends possible. The height is also around 160 cm tall, Friends is smaller than Kaleido, which is about 180 cm. We have to aim for a slimmer design, but we also have the pride of Kawasaki as a manufacturer of industrial robots, so we cannot make something that would soon break. We pursued a fine balance on Friends between the quality necessary for use in industry and a slimmer design. Furthermore, as we assume collaborative work with humans, the concept of collaborative robots is one of the features of Friends, such as not trapping a human’s hand as he or she gets close.
ー　The design is also one that will not make the people around the robot feel afraid.
Yes, that’s true. That is of course necessary for collaborative work with a human, but also because we believe that what people will demand of robots in the future will go beyond just labor. For example, when AI advances further, when you talk to “These flowers are pretty, don’t you think??but the AI cannot go out somewhere with you. However, a humanoid robot equipped with AI could live together with people. We believe that scenarios such as that might even become reality before use for labor. We made a particular effort for the acceptance of Friends by people, such as external appearance and facial expressions.
ー　Friends will be exhibited for the first time at iREX2022 (International Robot Exhibition 2022). Tell us the points to look out for at the exhibition.

We will showcase a practical application in a nursing care setting and will give a demonstration where Friends will push a wheelchair and have a conversation with an elderly person. Friends also play an entertainment performance where it will appear on the stage and dance with music.
The key technical point for this demonstration is the motion capture technology. A human shows a movement first and then the robot reproduces that movement.

Previously we programmed on a computer to operate a robot. However, this work is no longer necessary, as it will become possible to program intuitively. We chose dancing this time because we wanted people to see the movement of Friends, but teaching using motion capture can be used in a variety of situations.

Four-legged robot RHP Bex (hereinafter called “BEX?

ー　The RHP Bex is also something that will be revealed for the first time. Tell us the background of the development.
Through the development of Kaleido, we felt the difficulty of robots that walk on two legs. Because humanoid robots have the same form as humans, they have the possibility of eventually being able to do everything that humans can and having great versatility. However, it will probably still be a long time before we see their practical use. On the other hand, we are also developing the service robot Nyokkey that moves autonomously on wheels, and are able to move to the demonstration testing phase in about a year. It was possible to develop it in a short period of time, but when it comes to mobility over uneven ground, it is inevitably legs that are more appropriate than wheels. There must be an area in between a humanoid robot and Nyokkey. We thought that there may also be business opportunities in that area. It was for that reason that we started the development of the Bex robot that walks on four legs. We believe that the walking technology we have developed for humanoid robots will definitely also be applicable to four-legged robots.
ー　What kind of situations will Bex be used in?
First, we are assuming light payloads such as transporting materials in a construction site. We are currently conducting the development with the target of a 100 kg maximum payload. Another area is inspections. A possible method of use would be for Bex to patrol around the vast area of a plant and to check current state of measuring instruments by checking the camera images from a remote location. Another possible method of use would be for Bex to transport the crops that people have harvested on agricultural land. Kawasaki has positioned this Bex as one of the humanoid robots and is currently advancing its development. There are already four-legged robots, but there are not many robots with arms like a humanoid. We believe that this is the significance of working on four-legged robots at Kawasaki, which has been striving for the development of humanoid robots.

On the other hand, even though the base is a humanoid robot, the upper half of Bex is not fixed. We want to go on to adapt it to various applications. We will form partnerships and leave the upper half up to the other partners, which could be a construction company in the case of a construction site, or a plant manufacturer in the case of use in a plant. Kawasaki will focus efforts on the four legs that form the lower half of the body and we want to offer it as an open innovation platform.
ー　Tell us the points to look out for in the Bex exhibit at iREX.
The Kaleido weighs 80 kg. Its two legs support 80 kg. On the contrary, Bex has four legs, which are potentially capable of supporting 160 kg. We will demonstrate a human ride on the Bex. At iREX, come and see not only Kaleido, but also to its sibling Friends, and to Bex.

RELATED CONTENTS

]]> Kawasaki Robotics //treeforeurope.com/in/blog/story_22/ Thu, 09 Mar 2023 00:51:00 +0000 urn:uuid:98432b28-19df-402f-b47e-6b1227fcfced What is Kaleido?
“Kaleido?is the humanoid robot (human type robot) that Kawasaki is researching and developing. It has a high level of durability based on the technologies of Kawasaki, which has a history of over 50 years in the industrial robot sector. It has a physical size roughly equivalent to that of an adult human, but has a robust structure such that it does not break even if it falls over. We are aiming for its practical application in the future. It is 180 cm high and weighs 80 kg.

What is different about the 7th generation Kaleido?

ー　Tell us about the key points of the improved 7th generation Kaleido.
On the previous versions, we were first aiming to realize Kaleido walking on two legs. Although we succeeded in that, ordinary persons expect humanoid robots to move naturally in the same way as humans. We had not yet reached that level on the previous Kaleido. What we were pursuing in this update of Kaleido were the aspects of “walking and moving like a human? At the iREX this time, I would like the visitors to pay attention to the natural and quick movements of the RHP7.

ー　What does it mean specifically by walking and moving like a human?
In engineering theory, robots that walk on two legs can only walk while bending their knees. However, humans can walk easily with their knees straight. Also, humans land on their heels and push off with their toes to walk. This is just something we take for granted, but it is an operation that is very difficult to perform with the foot of a robot, which is made with a rigid body. Among these constraints, we succeeded in having Kaleido walk at a speed of 4 km per hour, which is the average walking speed of humans. If you watch the RHP7 walking, you should get a sense of how it walks more smoothly than before. The background to this achievement of quick walking is a technology called “dynamic behavior support? As the opposite of this, “static behavior support?is to ensure that when standing on two feet, the center of gravity is always in the center of the body. When walking, this must be done slowly to achieve balance. On the other hand, with “dynamic behavior support? it is possible to proactively perform movements so that the center of gravity shifts away from the leg that is supporting the body. The unstable state can be stabilized with the control to realize movements closer to that of humans.

Robustness to be able to fail

ー　Tell us the key points in the development.
First was what I just said about the control technology to stabilize the unstable state. In order to move in an unstable state, work is necessary to first predict what kind of instability will occur, and then to rapidly control that instability to stabilize. The technology required for that control is considerably advanced. Also, the development is a steady work with repeated failures. The robot falls over multiple times during the development process, so it is very important that the body does not break. From the beginning, the RHP name is an abbreviation of Robust Humanoid Platform. Robust means strong and sturdy. It has a robust structure that is unlikely to break, but even so, the body may suffer damage when repeating tests where the balance is actively disturbed. For this reason, it was necessary to make improvements for a body that was even more difficult to break.

For an open platform to improve humanoid technology

ー　Kaleido has come closer to human-like movements, so what are the things to watch out for at the iREX exhibition?
Our ultimate aim is the implementation of humanoid robots in society. To achieve this, we are asking what would happen if we had Kaleido perform work. Our exhibit assumes actual hazardous work. The first of these is work in a high place in a construction site. At present, this high-place work is performed by humans. Even with a lifeline, it is still work that is dangerous. We would like robots to perform that kind of work. In the exhibition this time, Kaleido will be hoisted up and perform work in a swinging state. In another exhibit, Kaleido goes up on a balance beam and maintains balance while walking. It jumps down from there. You can see the “dynamic behavior support?I just talked about in an easy way. From now on, I want Kaleido to go beyond Kawasaki to proceed the development and to be implemented in society in collaboration with many universities and companies. If we can cooperate with partners to further develop the humanoid robot technology as a platform, we welcome both Japanese and foreign parties. I would definitely like researchers to feel the potential of Kaleido through this “dynamic behavior support?

open-innovation space for robots
YouComeLab

Issues for robot development
There are demands for the implementation of robots in society for reasons such as labor force decline and the coronavirus crisis. On the other hand, there are various barriers to robot development, including the cost, auxiliary equipment, trouble support and repairs. Development is difficult at companies that are not robot manufacturers, and there have been issues such as that even if a robot manufacturer was requested to make a prototype, it was not possible to obtain the data for improvements.
YouComeLab　ー　A space where the development of new technology and new products can be conducted casually and easily with Kawasaki robots.
Kawasaki will establish YouComeLab as a venue for open innovation with institutions in various specialist areas, such as AI, machine learning, advanced sensors, advanced control, and new materials. The planned installations at YouComeLab include the humanoid robot Kaleido and other Kawasaki robots such as the remote collaboration system, the Successor. It will be possible for institutions such as companies and universities to use these robots to pursue their development themselves.
Merits of YouComeLab

Robot and equipment use Free of charge
During startup and troubleshooting Regularly stationed experts will give in-person lectures
Sophisticated systems Joint trials with Kawasaki

This has been in operation in our Akashi Works since April 2021. A new one will be opened in Haneda, Tokyo in April 2022, and another one in Nagakute, Nagoya in fiscal 2022. In the future, we also want to establish overseas bases such as in Silicon Valley, Beijing, Paris and elsewhere. Kawasaki will promote collaborations with many partners to realize the society where humans and robots co-exist and mutually prosper.

関連リンク

]]>