Redesigning a VR Laser Capture Tool: Boosting Human‑Machine Collaboration and Usability

With the rise of AI, cloud computing, VR and other new technologies, machines are gradually freeing or replacing human labor, creating new collaborative work modes and prompting fresh design challenges for user experience. This article uses 58.com’s self‑developed VR laser capture tool redesign project to illustrate design thinking for hardware‑software production tools in this context.

From Passive Machines to Collaborative Partners

Traditional design focused on issuing commands to machines, often ignoring the learning experience of inexperienced users. In the era of human‑machine collaboration, the focus shifts to "collaboration"—ensuring that human cognition and behavior synchronize with machines throughout the workflow. Designers must address the knowledge gap caused by new technologies and work processes.

The VR laser capture tool relies on automated algorithms to achieve fully automatic shooting and editing, collecting point‑cloud data and generating floor‑plan drawings. Its primary users are novice real‑estate agents with zero experience. Compared with traditional panoramic shooting, VR laser capture automatically builds spatial relationships, captures color and dimension data, and recreates three‑dimensional spaces.

Four Collaboration Modes

Research identifies four ways humans and machines cooperate:

Machine replaces human : machines perform repetitive tasks, e.g., washing machines saving time without simultaneous human involvement.

Human controls machine : humans make decisions while operating machines for dangerous or strenuous tasks, e.g., a driver operating a bulldozer.

Machine assists human : machines provide automation or suggestions to speed up human‑led work, e.g., Grammarly offering real‑time grammar corrections.

Human‑machine symbiosis : both understand each other's intent and guide decisions together, e.g., Siri learning user habits and providing personalized suggestions.

The VR laser tool currently sits in zones 1 & 2 but aims to transition toward zones 3 & 4, moving from passive to active machine participation.

Key Factors for Effective Collaboration

Understanding machine principles, sharing a common collaboration goal, ensuring smooth information exchange, and perceiving machine value are critical to efficient human‑machine interaction.

Experience differences can be examined from three layers:

Perception : users need to know machine functions, operation methods, status, and advantages.

Behavior : machines should actively participate, either replacing labor or assisting humans to achieve goals faster.

Cognition : users should grasp machine logic and shared objectives, while machines recognize user intent and offer decision support.

Usability Findings

Heuristic evaluation and a 4W analysis revealed major pain points:

Novice users face long, complex tasks with high learning and memory barriers, lacking awareness of machine collaboration.

Difficulty understanding point‑cloud and laser concepts and related terminology.

Insufficient interaction efficiency, delayed feedback, and poor error‑tolerant design.

Design Strategies

1. Contextual Teaching : Break new principles, tasks, and objects into small, situational learning nodes, using image cards and videos to build associative memory without overwhelming users.

2. Establish Collaboration Consensus : Video introductions clarify human‑environment‑device relationships; image cards state task goals, behavior norms, and how the machine assists, reinforcing a shared mental model.

3. Standardized Language : Define consistent stage names, task descriptions, and object labels using plain language, avoiding unfamiliar terms and ensuring smooth noun‑verb flow.

4. Scenario‑Based Allocation : Balance task priority between human and machine based on the machine’s capability level, e.g., “machine does – human checks/adjust” as the core workflow for the VR tool.

5. Efficient Information Transmission : Leverage digital‑twin visualizations to convey machine analysis and guide users to optimal shooting points.

6. Invisible Assistance : Automate repetitive steps, predict user intent, and intervene subtly without disrupting the user’s sense of control.

Results

After redesign, a usability test with 12 inexperienced users showed higher task completion rates, reduced time, and improved efficiency across shooting and editing phases, confirming the effectiveness of the redesign.

Future work will continue to validate the experience in real business scenarios, collect user feedback through flexible online/offline methods, and further refine the tool for easy learning and use.