Computational Design
HRI in Motion
The Project ‘HRI in Motion’, touches upon Human Robot Interaction and scenario-oriented Interface for navigational robots. Based on the research studies and robot developments from ReAC, we have designed a customized Bluetooth controlling system to provide workers in construction sites to request services from assistive robots in concise ways. The system is built up from two perspectives:Hardware and software. On the Hardware side, we have adapted two copies of interface: IOT keyboard devices as physical interface and Android APP as digital interface, both utilize Arduino as a middleware to establish bluetooth communication between the main PC of the robot and our remote devices. On the software side, we designed the communication system in a way that only utilized single digits instruction with recurring frequency to avoid latency with heavy loaded information transmission. With this strategy, we perform all the information storing, conversion and changes inside ROS to ensure the robustness for bluetooth communication. The final Android app is able to let workers in construction sites easily store customized parking locations for robots, command robots to parking locations, etc. The IOT keyboard device is also capable of these functions and allows communication over long distances. We also designed user-based UI graphic pages to let workers use the app without sophisticated instructions, so that people without prior knowledge of how the system works can still get their hands on the commands easily.
In the first state, prior to anycommands for the robot, workers will need to drive the robot to map the construction site. While completing this necessary task, workers can also mark locations that they want the robot to frequently visits, including but not limited to the following:
■ Charging station for the robot
■ Tool station for the robot
■ Material station or dumpster
■ Specific work zone/location forworkers to retrieve tools
When the robot enters working mode,it no longer requires to remember any locations, but will do navigation and path planning based on request. In our small demo scenario, we demonstrate how it will start from its charging station to tool loading station. After a worker loads tools onto the robot's tool box, it will then navigate to work zones and unload the tools, and finally go back to its charging station. All the dynamic path planning and collision avoidance pipelines for Husky are incorporated into the performance.
Using Heltec WiFi LoRa (ESP)32 as the communication board.This ESP32 microcomputer is equipped with a Long Range mesh network (LoRa) and able to transmit data in at least 100 meters range as we tested. It also has a Bluetooth module and WiFi module.
For the physical IOT demo please see how long distance navigation and command calls are enabled by the transmittance ability from Lora.
.png)
