Here is a framework to train quadruped robots using only RGB camera inputs. LucidSim creates diverse and physically accurate image sequences from the robot’s perspective. Training is done within simulated environments. Generative models are used to generate realistic visual data, enhancing the robot’s ability to generalize across different environments.
Meet the Ropet robot pet: an AI robot that can dance to your music and engage you in chats thanks to ChatGPT. It has a soft touch fur and maintains a body temperature similar to humans. This gorgeous robot has expressive eyes. It is smart enough to recognize over 100s of foods. It develops a unique personality as you interact with it.
As you interact with this robot, you earn points to unlock eye and voice effects. You can also customize its fur color. This robot has a high definition camera, stepper motors, speaker, and touch sensors. You can find this project on Kickstarter.
The Unitree G1 is one of the most impressive humanoid robots we have seen. It is very agile and can perform long jumps up to 1.4 meters. It uses a 3D LiDAR and depth camera to perceive its surroundings. Its operation data set is now open sourced. That includes learning algorithms, datasets, and models.
In the past few years, we have covered plenty of robots that can adapt to off-road conditions. The DEEPRobotics Lynx robot is one of the most impressive ones yet. This wheeled robot dog can do backflips, run up and down challenging terrains and even maintain its balance in bipedal mode. The below video shows it in action:
[HT]
In the past few months, we have seen plenty of robots that offer integration with large language models. While LLMs revolutionize robots with contextual reasoning and facilitate human-robot interaction, they open robots to risk of being jailbroken. RoboPAIR is an algorithm designed to jailbreak LLM-controlled robots. It “elicits harmful physical actions from LLM-controlled robots.” Here is what the researchers accomplished:
According to the researchers, in many scenarios, they managed to achieve 100% attack success rate. In the above video, you can see how the robot was tricked to deliver an explosive package.
[HT]
In the past few months, we have covered Agility Robotics’ Cassie robot doing a variety of tasks. David Ho, a master’s student in robotics at University of Michigan, has built a robot inspired by it. It stands 1ft tall and features inverted legs. To build this, David used a Raspberry Pi 4, 8 x 40kg and 2 x 80kg servos. He also used a MPU6050 IMU to measure orientation and body positin.
David is not done building robots. Apparently, he is considering developing a hopping kangaroo robot.
[HT]
The Eilik Robot is one of the cutest desktop robots we have tested. It has a bunch of fun mini games and can react to your touch. The Panxer is a RC battle tank for this robot. It gives your robot a mobile base to battle with friends. Its controller has a OLED screen that shows game information. This tank has dual vibration motors and sound effects. You can control it from a distance of up to 20 meters.
Those of you who have used solar panels in the past know the importance of pointing them to the sun to maximize the amount of energy they capture. The pSolBot 1 can help with that. It is a sun tracking robot for solar panels that uses solar mapping and real-time tracking to continuously make adjustments to point your panel towards the sun. It can capture 30-40% solar energy. This robot rests at night to conserve energy.
Here is another humanoid robot that will automate tasks in various settings, including factories. The XPENG Iron is a 5-foot-10-inch robot, weighing 70kg. It has an AI brain that runs on a Turing AI chip. It has 60 joints and 200 degrees of freedom. The below video gives you a better idea how this robot works:
Known for following in Tesla’s footsteps, XPENG is doing so quite literally with humanoid robots.
Today, XPENG announced Iron, a 5-foot-10-inch robot that weighs 153 lb (70 kg) and has human-like dexterous hands.
⦿ Iron robot has begun ‘working’ in their vehicle factory in… pic.twitter.com/tqGiA6yPJ9
— The Humanoid Hub (@TheHumanoidHub) November 6, 2024
[HT]
