Playground

The Playground is a powerful interactive environment provided by the Ticos platform where you can test and converse with your created Embodied Agents in real-time. It simulates the agent’s behavior on real-world terminals, allowing you to comprehensively evaluate its functionality, responses, and user experience before deployment.

Prerequisites

Before you begin using the Playground, ensure you have met the following requirements:

1. Created an Agent: You must have at least one Embodied Agent created and configured under your account. If you don’t have one yet, please go to the Agent Configuration module to create one.
2. Available Online Terminal: You must have at least one terminal device registered to your account and in an “online” status. If the dropdown menu shows “No terminals found”, please go to the Terminal Ops module to check your terminal status or register a new terminal.

Interface Layout

The Playground page is divided into three main areas:

1. Control and Selection Area

Located at the top of the page, containing core control functions:

• Select Agent: A dropdown menu listing all available Embodied Agents under your account. You must select an agent to interact with here.
• Select Terminal: A dropdown menu listing all currently online terminals available for interaction.
• Start/End Chat: A core button used to initiate or terminate an interactive session with the agent.

2. Perception and Status Area

The main part of the page, displaying real-time feedback from the terminal and the agent’s internal state:

• Perception Feedback:
  • Face Detection: Shows whether the camera detects any faces.
  • Object Detection: Shows objects identified by the camera.
• Agent Status:
  • Current Emotion: Displays the agent’s current emotional state in both graphical and textual form (e.g., idle, happy, sad).
  • Current Action: Shows the action or posture the agent is currently performing (e.g., default, waving).

3. Interaction and Log Area

Located on the right side of the page, containing two tabs for monitoring conversations and system events:

• Messages: Displays the complete conversation history between you and the agent. You can enter text here to communicate with the agent.
• Events: A real-time log stream showing important events generated by the system in the background, such as function calls, state changes, error messages, etc. This is very useful for debugging.

Usage Process

1. Navigate: Access the “Playground” from the main menu on the left.
2. Select Agent: Choose the agent you want to test from the dropdown menu at the top.
3. Select Terminal: Select an online terminal to run the agent.
4. Start Chat: Click the “Start Chat” button, and the system will establish a connection, making the interactive environment ready.
5. Interact: Begin conversing with your agent. Observe its responses as well as changes in the perception and status area.
6. Monitor and Debug: View detailed system logs in the “Events” tab to understand the agent’s internal decision-making process.
7. End Session: After completing the test, click the “End Chat” button to disconnect.

Advanced Settings

Click the settings icon button on the right side of the top action area to enter the advanced settings page, where you can configure specific options for the experience functionality.

Notes

• If the “Select Terminal” dropdown is empty or shows “No terminals found”, please confirm that you have created terminal devices and their status is shown as “Online” on the “Terminal Ops” page.
• The agent’s behavior is entirely dependent on the instructions, knowledge, skills, etc. that you set for it in the “Agent Configuration” module. If you find behavior that does not meet expectations during the experience, please return to the configuration page for adjustments.