Robotic Sensor Identifies Shapes of Crops Using Sound

Researchers from Carnegie Mellon University’s Robotics Institute (RI) invented a tool called SonicBoom that can find crops like apples based on the sound they make. The research team used an array of six contact microphones placed inside a piece of PVC pipe. When the pipe touches an object, such as a tree branch, the microphones detect the resulting vibration. By analyzing the differences in the sound waves, the researchers were able to triangulate where the contact took place. SonicBoom can localize contacts with a precision between 0.43 and 2.2 centimeters.

The PVC pipe protects the contact microphones from damage. It also gives the appearance of a microphone boom, inspiring the name SonicBoom. Ultimately, the microphones could be installed inside a robot arm. The researchers used a data-driven machine learning module to develop the ability to map the signals from the microphones. To do so, they collected audio data from 18,000 contacts between the sensor and a wooden rod. Using the audio data, SonicBoom determines the location of hard or rigid objects. Changing its configuration should enable it to also sense less rigid objects, such as soft fruits and vegetables.

More information:

https://www.cmu.edu/news/stories/archives/2025/august/new-robotic-agricultural-sensor-could-revolutionize-farming

Robot Mall

The world's first Robot Mall opened in Beijing's Yizhuang and has a total area of about 4,000 square meters across four floors, showcasing seven major categories and more than 50 robot brands, covering various fields including medical, industrial, bionic, and humanoid robots. Visitors can interact with various types of robots, including robot dogs and robot hands, as well as compete with robot athletes in football, basketball and chess.

There are showcases of a one-stop service process for robot part replacement, diagnostics, and remote operation and maintenance, and a high-end negotiation space to facilitate precise industry matchmaking and support business cooperation. In 2024, China accounted for two-thirds of global robot patent applications and produced 556,000 industrial robots, remaining the world's top manufacturer.

More information:

https://www.globaltimes.cn/page/202508/1340351.shtml

World Humanoid Robot Games

China kicked off the three-day long World Humanoid Robot Games on Friday, looking to showcase its advances in artificial intelligence and robotics with 280 teams from 16 countries. Robots competed in sports such as track and field, and table tennis, as well as tackled robot-specific challenges from sorting medicines and handling materials to cleaning services.

Teams came from countries including the United States, Germany and Brazil, with 192 representing universities and 88 from private enterprises such as China's Unitree and Fourier Intelligence. Competing teams used robots from Chinese manufacturers such as Booster Robotics.

More information:

https://www.reuters.com/sports/robots-race-play-football-crash-collapse-chinas-robot-olympics-2025-08-15/

Smartphone Audio Guide for Visual Impaired

Tactile paving for people with visual impairments in Japan has recently been enhanced with audio guidance, as the developers of the technology aim to improve the lives of people with partial sight and assist sighted tourists. The new braille blocks are marked with black stickers in special patterns. By scanning the coded blocks with a smartphone camera and app, users can listen to audio information about the location and its surroundings. The system was jointly developed by the laboratory of Kanazawa Institute of Technology and Tokyo-based W&M systems LLC.

Those behind the technology hope it can also be made available to tourists and foreigners as they walk the streets of Japan, whether they are sighted or not. The special braille blocks were first introduced in Kanazawa, central Japan, in 2019 and have since been installed in other areas, appearing at train stations, pedestrian streets and public offices in 10 prefectures, including Tokyo and Osaka, as of April. The developers plan to make the system available in multiple languages and are considering enabling it to answer questions by incorporating generative artificial intelligence capabilities.

More information:

https://japantoday.com/category/national/speaking-tactile-sidewalks-enrich-lives-of-visually-impaired

Generated Video Teaches Robots

University of Illinois Urbana-Champaign researchers developed a video generation model to produce demonstration videos to teach robots to perform specific tasks. Users provide a text command and an image of the current scene to produce a demonstration video tailored to the robot's specific task and environment, with no additional training required. RIGVid's success rate across tasks was 85%, outperforming other robot imitation methods.

RIGVid’s approach for extracting an object’s motion from a generated video, based on model-based six DoF object pose tracking, also outperformed alternative methods for trajectory extraction using optical flow and sparse keypoints. The improvement was apparent on challenging tasks, like sweeping dirt or placing a thin spatula into a pan, where the other methods struggled with object occlusion.

More information:

https://siebelschool.illinois.edu/news/RIGVid