Undoubtedly, VSLAM will be one of the keys to promoting the robot to iterate again.
Moving "machines" or robots?
As the number of landing robots continues to grow, the product defects that were “ignored” in the early stage are causing more and more serious negative impacts. On the one hand, although laser SLAM is constantly evolving based on the Cartographer algorithm, there has been almost no significant progress in recent years. Due to the fact that laser can only build a pure geometric environment model, it cannot "see" the real environment and lacks the ability to interpret the content of the environment. There are natural bottlenecks in the realization of intelligent interaction, intelligent obstacle avoidance, and intelligent operation, which greatly limits the intelligent upgrade of robots. As end users pay more and more attention to intelligence, coupled with the emergence of AI large models, it is inevitable to upgrade new navigation technologies.
On the other hand, in terms of deployment issues, the traditional marker positioning and laser positioning technology solutions not only have a long deployment cycle, high cost, and complicated operations, but also require the support of an offline engineer team, and have poor environmental adaptability. Scenarios often require environmental modification, which further increases the difficulty of deployment. At the same time, once the environment layout changes, it needs to be re-deployed, which greatly increases the cost of use. In addition, the high cost of high-precision lidar has also indirectly led to the high price of robots, and the artificial replacement value is not obvious.
An expensive robot not only needs long-term assistance and support from a team of engineers, but also is not relatively intelligent. Obviously, it is more defined as a mobile "machine" than a robot.
The meaning of VSLAM is not just low price
Under the multiple pressures of the market and the industry, leading manufacturers have been trying new technical solutions. With the rise of VSLAM technology, people unanimously foresee the huge imagination behind it.
Humans can obtain external information through senses such as hearing, touch, and smell, but more than 80% of it comes from vision. Through vision, the human body can obtain information such as the size, position, light and shade, color, posture, etc. of external objects, so as to complete the interaction with the surrounding environment. The robot obtains external information by means of various sensors: from the earliest collision feedback, to lidar scanning, and then to the stereoscopic vision system based on "human eye" bionics composed of dual (multi) cameras, completing the transformation from one-dimensional To two-dimensional, and then to three-dimensional upgrade.
Through bionic human vision, the robot has acquired the perception of space and has the basic ability and premise of "human". This is the biggest difference between VSLAM and other technical solutions, and it is also its core meaning and value.
At the same time, the natural cost advantage of the visual sensor can reduce the cost of the visual solution by at least 50% while achieving the same technical effect as the laser solution, which also means that it can pay back the cost faster.
According to calculations by industry insiders, in commercial cleaning robots, navigation BOM costs account for about 40%, and the robot’s selling price is calculated at 50% gross profit margin. The BOM cost of the VSLAM solution for a navigation module that meets the application accuracy of commercial scenarios is about 6,000 yuan. The price of the machine is about 30,000 yuan, the BOM cost of the laser SLAM solution is about 12,000 yuan, and the price of the whole machine is about 60,000 yuan. Based on the 3-year service life, the payback time of the visual SLAM solution is about 10 months, with an average annual ROI of about 87%. The payback time of the laser SLAM solution is about 1 year and 8 months, with an average annual ROI of about 27%.
Obviously, the VSLAM solution will have a better chance of bringing about an inflection point where robot costs are lower than labor costs.
It can be seen that some leading manufacturers on the market have shifted their technical routes to VSLAM. For example, Purdue has self-developed the PUDU VSLAM+ solution, which is a new generation of VSLAM technology developed based on visual cameras, and has been applied to its "Happy Send 2" robot superior. And INDEMIND, as the first batch of AI technology companies in China to deploy computer vision, is also at the forefront of the industry in visual research and development.
Since its establishment, INDEMIND has insisted on cultivating visual navigation, and has accumulated deep accumulation and experience in key technologies and product development such as robot navigation, obstacle avoidance, decision-making, and AI interaction. The core commercial robot navigation solution "Commercial Robot AI Kit" has reached the industry-leading level in terms of overall performance.
More mature, smarter, "a generation ahead" experience
"Commercial Robot AI Kit" is based on INDEMIND's self-developed OS Fusion AGI system, adopts a multi-sensor fusion architecture centered on a binocular stereo camera, supports different types of mainstream sensors on the market, and satisfies commercial robot navigation and positioning, intelligent avoidance The development of core functions such as obstacles, path planning, and decision-making interaction can be widely used in commercial service robot platforms such as commercial cleaning, hotel delivery, meal delivery, and inspection.
In terms of applicability, the "Commercial Robot AI Kit" is almost fully compatible with commercial scenarios, and has an effective solution for highly transparent materials such as glass. At present, it has been deployed in thousands of different commercial scenarios around the world without scene classification. The built-in active supplementary light strategy can work without difference in special lighting environments such as strong direct light, no light source, and dim light, meeting the requirements of all-weather operations.
At the same time, INDEMIND has specially developed an intelligent decision-making engine, which effectively improves the robot's decision-making ability. Based on the powerful perception & cognitive ability brought by stereo vision technology, it can realize various key functions such as intelligent obstacle avoidance, active safety, intelligent operation, intelligent interaction, intelligent ladder control, etc. Taking intelligent interaction as an example, based on the three-dimensional space information and The 3D map and AI recognition function of object semantic information, combined with the intelligent decision-making engine, can call the robot by voice control, let the robot follow, specify the target area to work, etc., greatly improving the intelligent experience of the robot.
In terms of cost, the "Commercial Robot AI Kit" is based on the fusion navigation solution with the visual sensor as the core and the advantages of the INDEMIND OS Fusion system. The lowest development cost of the robot can be lowered to less than 2,000 yuan, and the cost of the complete chassis including navigation and battery can be Dropping to less than 5,000 yuan, compared with the lidar solution, it has dropped by 60-80%, which significantly reduces the development cost and cycle of the robot.
In addition, the robot equipped with the "commercial robot AI Kit" can build maps independently in real time, so that new machines and new scenes do not need to be pre-deployed, and can be used out of the box. After changing the scene, the robot can also update the map independently without relying on offline Support team, remote areas can completely rely on network sales.
At present, INDEMIND has reached cooperation with many giant customers at home and abroad. The medium-sized cleaning robot Cobi18, which cooperates with traditional cleaning manufacturer ICE, has been deployed in batches in more than a dozen countries around the world, and has achieved zero failure in European and American markets, and has received extensive praise and feedback. In the future, INDEMIND expects orders to exceed 100,000 units in three years.